video_id
stringlengths
59
122
avg_aesthetic_score_laion_aesthetics
float64
2.78
6.46
frame_aesthetic_scores_laion_aesthetics
stringlengths
72
273
video_name
stringlengths
43
106
video_path
stringlengths
86
149
motion_fb_motion_score
float64
-13.69
24.8
motion_lk_motion_score
float64
0
9.76
frame_shot_categorization_shot_categorizer
stringlengths
276
1.91k
avg_subjects_subject_presence
float64
0
39.3
avg_humans_subject_presence
float64
0
26
avg_animals_subject_presence
float64
0
25.3
frame_counts_subject_presence
stringlengths
268
1.61k
avg_vision_reward
float64
-0.15
0.15
frame_wise_rewards
stringlengths
78
104
video_url
stringlengths
39
39
scene_name
stringclasses
79 values
SCENE_NUM
float64
1
79
START_FRAME
float64
24
13.7k
END_FRAME
float64
73
13.8k
START_TIMECODE
stringlengths
12
12
END_TIMECODE
stringlengths
12
12
START_SECONDS
float64
1
570
END_SECONDS
float64
3.04
574
DURATION_SECONDS
float64
2
6.97
GAP_FROM_PREVIOUS
float64
1
219
prompt_x
stringclasses
1 value
caption_t2v_style
stringlengths
31
2.47k
prompt_y
stringclasses
1 value
caption_t2v_style_short
stringlengths
34
3.66k
duration_seconds
float64
2
6.97
width
int64
640
1.92k
height
int64
262
1.08k
total_frames
int64
48
201
resolution
stringclasses
238 values
short_scene_001_short_$20_Bill_-_Paper_Moon_(5_8)_Movie_CLIP_(1973)_HD
4.908089
{'frame_00:00.png': 4.874268531799316, 'frame_00:01.png': 5.056090354919434, 'frame_00:02.png': 4.996335506439209, 'frame_00:03.png': 4.955479145050049, 'frame_00:04.png': 4.690274715423584, 'frame_00:05.png': 4.842457294464111, 'frame_00:06.png': 4.941718578338623}
short_$20_Bill_-_Paper_Moon_(5_8)_Movie_CLIP_(1973)_HD
../assets/split_scenes/short_$20_Bill_-_Paper_Moon_(5_8)_Movie_CLIP_(1973)_HD/short_scene_001.mp4
0.283501
0.800726
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:05.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:06.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
7.857143
6.571429
0.142857
{"0": {"raw_counts": {"person": 7, "tie": 1, "total": 8}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 8}}, "1": {"raw_counts": {"person": 6, "tie": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 7}}, "2": {"raw_counts": {"person": 7, "tie": 2, "total": 9}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 9}}, "3": {"raw_counts": {"person": 7, "tie": 1, "total": 8}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 8}}, "4": {"raw_counts": {"person": 8, "tie": 1, "horse": 1, "total": 10}, "bucketed_counts": {"humans": 8, "animals": 1, "vehicles": 0, "food": 0, "others": 1, "total": 10}}, "5": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "6": {"raw_counts": {"person": 5, "chair": 2, "total": 7}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 7}}}
0.018693
[-1, 1, 1, -1, -1, 1, 1, -1, -1, 1, -1, 1, 1, -1, 1, -1, -1, -1, 1, -1, -1, 1, 1, -1, 1, 1]
https://youtube.com/watch?v=CKJJbZe4TWM
scene_001
1
55
217
00:00:02.294
00:00:09.051
2.293961
9.050717
6.756757
2.293961
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The black and white shot depicts a man in a pinstripe suit and fedora standing at a store counter, pulling a wallet from his jacket and preparing to pay. The shot is composed with the man in the center left, an ornate metal cash register between him and the store clerk, whose back is to the camera. Hats and hat boxes line the shelves behind the clerk and above the man. To the left and back, a woman with a cloche hat browses other hats, and further back in the shop, other customers and staff are visible. A bright round light fixture hangs from the ceiling above the counter. Another clerk stands to the right, wrapping a purchase. The scene is filled with details of a 1930s-era shop.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man wearing a suit and hat pays a store clerk at a vintage cash register in a bustling shop during the day. Shoppers and other employees are visible in the background amidst displays of merchandise.
6.756757
1,280
540
162
1280x540
short_scene_001_short_(500)_Days_of_Summer_(2_5)_Movie_CLIP_-_Playing_House_(2009)_HD
4.640624
{'frame_00:00.png': 4.344003677368164, 'frame_00:01.png': 4.3860321044921875, 'frame_00:02.png': 4.786259651184082, 'frame_00:03.png': 4.876478672027588, 'frame_00:04.png': 4.8103437423706055}
short_(500)_Days_of_Summer_(2_5)_Movie_CLIP_-_Playing_House_(2009)_HD
../assets/split_scenes/short_(500)_Days_of_Summer_(2_5)_Movie_CLIP_-_Playing_House_(2009)_HD/short_scene_001.mp4
0.004516
0.069386
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
2.6
1.6
0.2
{"0": {"raw_counts": {"person": 2, "tv": 2, "laptop": 1, "total": 5}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 5}}, "1": {"raw_counts": {"cat": 1, "tv": 1, "total": 2}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.082482
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 1 -1 1 1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=ARoB1nWPsxo
scene_001
1
470
590
00:00:19.603
00:00:24.608
19.602936
24.607941
5.005005
19.602936
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man and woman are sitting on a couch facing a blank television screen. The television is centered in the frame with the backs of the two people on either side. The man’s face is partially visible, while the woman is angled to the side. Her face is fully visible. There is a sign on the right side of the frame that says “Living Room.” The scene transitions to a close-up of the two people. They appear to be having a conversation. The scene is brightly lit.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A couple is sitting together on a sofa in front of a television. The woman turns to look at the man beside her.
5.005005
1,920
790
120
1920x790
short_scene_001_short_(500)_Days_of_Summer_(3_5)_Movie_CLIP_-_I_Love_Us_(2009)_HD
5.191617
{'frame_00:00.png': 5.222700119018555, 'frame_00:01.png': 5.225332736968994, 'frame_00:02.png': 5.145378112792969, 'frame_00:03.png': 5.17305850982666}
short_(500)_Days_of_Summer_(3_5)_Movie_CLIP_-_I_Love_Us_(2009)_HD
../assets/split_scenes/short_(500)_Days_of_Summer_(3_5)_Movie_CLIP_-_I_Love_Us_(2009)_HD/short_scene_001.mp4
0.096495
0.055965
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
3.5
1.5
0
{"0": {"raw_counts": {"person": 2, "chair": 1, "laptop": 1, "clock": 1, "total": 5}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 5}}, "1": {"raw_counts": {"person": 1, "chair": 1, "laptop": 1, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 3}}, "2": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "3": {"raw_counts": {"person": 2, "chair": 1, "laptop": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}}
0.042632
[-1 1 1 1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 -1 1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=4rGia6hIWmk
scene_001
1
167
241
00:00:06.965
00:00:10.052
6.965299
10.051718
3.08642
6.965299
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person sits at a desk, facing another person whose back is to the camera. The person at the desk is wearing a white shirt, a dark tie, and a tan sweater vest. They have a serious, contemplative expression. Behind them on the wall are three framed drawings of cherubs in various poses. The first cherub is holding a bow and arrow, the second is floating, and the third is wearing a top hat. The scene is dimly lit, and there is a sense of tension or seriousness in the air. A mug is visible on the desk in front of the second person.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man wearing a sweater vest sits at a desk in an office setting and stares forward with a solemn expression. Behind him are three cartoon depictions of cherubs.
3.08642
1,920
810
74
1920x810
short_scene_001_short_(500)_Days_of_Summer_(5_5)_Movie_CLIP_-_All_That_True_Love_Nonsense_(2009)_HD
5.509413
{'frame_00:00.png': 5.444409370422363, 'frame_00:01.png': 5.392215251922607, 'frame_00:02.png': 5.636802673339844, 'frame_00:03.png': 5.564223766326904}
short_(500)_Days_of_Summer_(5_5)_Movie_CLIP_-_All_That_True_Love_Nonsense_(2009)_HD
../assets/split_scenes/short_(500)_Days_of_Summer_(5_5)_Movie_CLIP_-_All_That_True_Love_Nonsense_(2009)_HD/short_scene_001.mp4
-0.123754
0.128749
{'frame_00:00.png': [{'<COLOR>': 'Cool, Green'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}]}
2.25
2
0.25
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "dog": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
0.063299
[ 1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 -1 -1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=bBLKvPSgQ2A
scene_001
1
518
603
00:00:21.605
00:00:25.150
21.604938
25.15015
3.545212
21.604938
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with dark hair and bangs gazes at a man next to her, a slight smile on her face and earrings visible, while the man looks to his left, a hint of a smile playing on his lips, with green grass blurred in the background suggesting an outdoor setting. As she speaks, her smile widens, revealing her teeth and intensifying the friendly, pleasant atmosphere of their conversation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with dark hair smiles and looks toward a man with short hair who is turned toward her in a park setting, both are engaged in a conversation. The mood seems intimate and friendly, with a soft, natural background.
3.545212
1,920
810
85
1920x810
short_scene_001_short_10,000_BC_(7_10)_Movie_CLIP_-_We_Hunt_Together_(2008)_HD
4.335912
{'frame_00:00.png': 3.9596991539001465, 'frame_00:01.png': 4.329439163208008, 'frame_00:02.png': 4.718596458435059}
short_10,000_BC_(7_10)_Movie_CLIP_-_We_Hunt_Together_(2008)_HD
../assets/split_scenes/short_10,000_BC_(7_10)_Movie_CLIP_-_We_Hunt_Together_(2008)_HD/short_scene_001.mp4
-0.001026
1.377055
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Moonlight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Moonlight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Moonlight'}, {'<COMPOSITION>': 'Balanced'}]}
2.333333
1.666667
0.666667
{0: {'raw_counts': {'elephant': 1, 'person': 2, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 1, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 3}}, 1: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 2: {'raw_counts': {'person': 2, 'elephant': 1, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 1, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 3}}}
-0.107104
[-1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=zGrIGZifpwg
scene_001
1
41
97
00:00:01.710
00:00:04.046
1.710043
4.045712
2.335669
1.710043
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men stand side by side, their faces close to the camera, both with dark skin and some sort of white objects protruding from their lower lips, and their bald heads are partially illuminated. They are surrounded by others, shrouded in darkness. They stand still, their expressions suggesting tension or solemnity, the dark lighting adding an air of mystery to the scene. The shot is captured in close-up, emphasizing the minute details of their faces and the textures of their garments. The composition suggests a sense of solidarity or shared experience among the figures.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Tribal warriors stand guard, their faces adorned with bone-like ornaments, amidst a dark and mysterious tribal ritual scene. The dramatic lighting creates an intense, foreboding atmosphere.
2.335669
1,920
810
56
1920x810
short_scene_001_short_10_Cloverfield_Lane_(2016)_-_A_Narrow_Escape_Scene_(7_10)_|_Movieclips
4.091988
{'frame_00:00.png': 4.223435401916504, 'frame_00:01.png': 4.022943019866943, 'frame_00:02.png': 4.0295867919921875}
short_10_Cloverfield_Lane_(2016)_-_A_Narrow_Escape_Scene_(7_10)_|_Movieclips
../assets/split_scenes/short_10_Cloverfield_Lane_(2016)_-_A_Narrow_Escape_Scene_(7_10)_|_Movieclips/short_scene_001.mp4
-0.000772
2.557929
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
0
0
0
{0: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 1: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 2: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}}
-0.147119
[-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1]
https://youtube.com/watch?v=bMjYOV8VnYk
scene_001
1
191
244
00:00:07.966
00:00:10.177
7.9663
10.176844
2.210544
7.9663
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In an environment enveloped by darkness, a close-up shot reveals a face partially obscured by strands of hair, capturing a sense of mystery. The light illuminates a section of the mouth, creating a stark contrast against the deep shadows. As the scene progresses, a long, slender object, possibly a knife or a similar sharp tool, enters the frame from the periphery, approaching the individual.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A figure with disheveled hair emerges from darkness, with dramatic lighting highlighting their face. The scene creates a suspenseful and unsettling atmosphere.
2.210544
1,920
810
53
1920x810
short_scene_001_short_10_Cloverfield_Lane_(2016)_-_Everyone_Outside_Is_Dead_Scene_(2_10)_|_Movieclips
5.321013
{'frame_00:00.png': 5.081283092498779, 'frame_00:01.png': 5.472167015075684, 'frame_00:02.png': 5.4095892906188965}
short_10_Cloverfield_Lane_(2016)_-_Everyone_Outside_Is_Dead_Scene_(2_10)_|_Movieclips
../assets/split_scenes/short_10_Cloverfield_Lane_(2016)_-_Everyone_Outside_Is_Dead_Scene_(2_10)_|_Movieclips/short_scene_001.mp4
0.076658
0.079875
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
1
1
0
{0: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 1: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 2: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}}
0.034188
[-1 1 1 1 -1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=jypAc6XYFfA
scene_001
1
42
98
00:00:01.752
00:00:04.087
1.751752
4.087421
2.335669
1.751752
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with tears in her eyes sits with her back against a cinder block wall, listening intently to someone who is standing just out of frame to the left. She is wearing a white tank top. The lighting is dim and low-key, casting shadows that emphasize her worried expression. The drab texture of the wall heightens the somber mood.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with tears in her eyes sits slumped against a concrete wall as she stares at an indistinct figure standing before her. The scene is dimly lit and evokes a somber mood.
2.335669
1,920
810
56
1920x810
short_scene_001_short_10_Cloverfield_Lane_(2016)_-_Frank_&_Mildred_Scene_(3_10)_|_Movieclips
4.949686
{'frame_00:00.png': 4.909789562225342, 'frame_00:01.png': 4.878175258636475, 'frame_00:02.png': 5.061092376708984}
short_10_Cloverfield_Lane_(2016)_-_Frank_&_Mildred_Scene_(3_10)_|_Movieclips
../assets/split_scenes/short_10_Cloverfield_Lane_(2016)_-_Frank_&_Mildred_Scene_(3_10)_|_Movieclips/short_scene_001.mp4
0.007832
0.023206
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Cyan'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Cyan'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Cyan'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}]}
0
0
0
{0: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 1: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 2: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}}
-0.038117
[-1 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=Mc_zp1s60NY
scene_001
1
204
264
00:00:08.509
00:00:11.011
8.508509
11.011011
2.502503
8.508509
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A dark, cylindrical tunnel with ribbed metal walls curves inward towards a set of steps leading up to a bright, square opening at the end. A metal railing runs alongside the steps, providing a sense of guidance through the shadowy space. The scene is filled with darkness, with the only light source emanating from the opening, casting long shadows and highlighting the texture of the tunnel walls. The steps appear to be the only way out, creating a sense of anticipation and mystery.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A dimly lit tunnel leads to a set of stairs with a bright light at the top. The scene evokes a sense of mystery and anticipation.
2.502503
1,920
810
60
1920x810
short_scene_001_short_10_Cloverfield_Lane_(2016)_-_I_Accept_Your_Apology_Scene_(5_10)_|_Movieclips
4.147485
{'frame_00:00.png': 4.036964416503906, 'frame_00:01.png': 4.023606300354004, 'frame_00:02.png': 4.233264446258545, 'frame_00:03.png': 4.296105861663818}
short_10_Cloverfield_Lane_(2016)_-_I_Accept_Your_Apology_Scene_(5_10)_|_Movieclips
../assets/split_scenes/short_10_Cloverfield_Lane_(2016)_-_I_Accept_Your_Apology_Scene_(5_10)_|_Movieclips/short_scene_001.mp4
0.175419
0.055552
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1
0.5
0
{0: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 1: {'raw_counts': {'vase': 1, 'person': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 2: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 3: {'raw_counts': {'surfboard': 1, 'total': 1}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 1}}}
-0.085315
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=DJTF3NmqF7U
scene_001
1
130
218
00:00:05.422
00:00:09.092
5.422089
9.092426
3.670337
5.422089
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in a plaid shirt opens a large blue barrel. The gloved hand lifts the white lid, revealing a dark liquid inside. The scene focuses closely on the gloved hand and the blue barrel, emphasizing the opening action. The plaid shirt suggests a casual, possibly clandestine setting. The dark liquid inside the barrel creates an atmosphere of mystery or potential danger, drawing attention to what might be concealed within.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person wearing gloves lifts the lid of a blue barrel, revealing a dark liquid inside. The scene is dimly lit, creating a mysterious atmosphere.
3.670337
1,920
810
88
1920x810
short_scene_001_short_127_Hours_(3_3)_Movie_CLIP_-_Radio_Show_Breakdown_(2010)_HD
4.728652
{'frame_00:00.png': 4.68841552734375, 'frame_00:01.png': 4.7227630615234375, 'frame_00:02.png': 4.756557464599609, 'frame_00:03.png': 4.74687385559082}
short_127_Hours_(3_3)_Movie_CLIP_-_Radio_Show_Breakdown_(2010)_HD
../assets/split_scenes/short_127_Hours_(3_3)_Movie_CLIP_-_Radio_Show_Breakdown_(2010)_HD/short_scene_001.mp4
-0.308496
0.076074
{'frame_00:00.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}]}
1
1
0
{0: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 1: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 2: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 3: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}}
-0.039919
[-1 1 1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=6KqG8CYcMKk
scene_001
1
502
589
00:00:20.938
00:00:24.566
20.937604
24.566233
3.628629
20.937604
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds with a man positioned at a low angle, gazing upwards. His expression is etched with intensity, and he wears a cap. The backdrop features a textured, rocky surface, implying he's in a confined or natural setting. As the scene progresses, his mouth opens wider, suggesting a cry or expression of intense emotion. The framing remains tight on his face, emphasizing his reaction. A small portion of another person's gear is visible on the left side of the frame, hinting at the presence of a companion or possible danger.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man trapped between rocks looks upwards with a distressed expression, while another figure stands over him. The scene is lit with natural light reflecting off the rock walls.
3.628629
1,920
874
87
1920x874
short_scene_001_short_12_Angry_Men_(1_10)_Movie_CLIP_-_Kids_These_Days_(1957)_HD
4.964292
{'frame_00:00.png': 4.93519926071167, 'frame_00:01.png': 5.0059661865234375, 'frame_00:02.png': 4.951711654663086}
short_12_Angry_Men_(1_10)_Movie_CLIP_-_Kids_These_Days_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(1_10)_Movie_CLIP_-_Kids_These_Days_(1957)_HD/short_scene_001.mp4
0.275389
0.034144
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
8.333333
6
0
{0: {'raw_counts': {'person': 6, 'tie': 2, 'total': 8}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 2, 'total': 8}}, 1: {'raw_counts': {'person': 6, 'tie': 3, 'total': 9}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 3, 'total': 9}}, 2: {'raw_counts': {'person': 6, 'tie': 2, 'total': 8}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 2, 'total': 8}}}
0.007568
[-1 1 1 -1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=46kWbFAMHoc
scene_001
1
274
345
00:00:11.428
00:00:14.389
11.428095
14.389389
2.961295
11.428095
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The black and white image shows a group of men gathered around a table, engaged in a serious discussion. The composition is focused on the faces and upper bodies of the men. Two men are seated closest to the viewer, one on the left with only his shoulder visible, and another on the right wearing a dark jacket and tie. The man in the center, dressed in a light-colored suit and tie, looks directly at the viewer. Behind them, another man wearing a shirt and tie is visible, appearing slightly out of focus. The men's expressions are intense and focused, suggesting a critical or tense situation. The setting appears to be a room with reflective surfaces or windows in the background. The overall atmosphere conveys a sense of gravity and deliberation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of men are gathered at a table in what appears to be a meeting or deliberation, with some looking intently while others seem concerned or thoughtful. The scene is shot in black and white, which gives it a classic cinematic feel.
2.961295
1,920
810
71
1920x810
short_scene_001_short_12_Angry_Men_(2_10)_Movie_CLIP_-_It's_the_Same_Knife!_(1957)_HD
4.885741
{'frame_00:00.png': 4.868114948272705, 'frame_00:01.png': 4.835174083709717, 'frame_00:02.png': 4.950646877288818, 'frame_00:03.png': 4.868897914886475, 'frame_00:04.png': 4.905872821807861}
short_12_Angry_Men_(2_10)_Movie_CLIP_-_It's_the_Same_Knife!_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(2_10)_Movie_CLIP_-_It's_the_Same_Knife!_(1957)_HD/short_scene_001.mp4
-0.020022
0.23188
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
3.4
2.4
0
{0: {'raw_counts': {'person': 2, 'tie': 1, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 3}}, 1: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 2: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 3: {'raw_counts': {'person': 2, 'tie': 1, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 3}}, 4: {'raw_counts': {'person': 2, 'tie': 1, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 3}}}
0.006052
[-1 1 1 -1 -1 1 -1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=TUzp2XUhskY
scene_001
1
359
479
00:00:14.973
00:00:19.978
14.973307
19.978312
5.005005
14.973307
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene shows three men sitting in a room, seemingly engaged in a serious discussion. The man on the right, dressed in a light-colored suit and dark tie, looks intently at the person across from him, appearing thoughtful and slightly concerned. On the table between them is a knife. The man across from him is only partially visible from the back, also wearing a suit, suggesting a formal or professional setting. Another man sits partially in the background, also wearing a suit. The room is decorated with paneled walls and a closed door. The scene is shot with dramatic lighting, enhancing the tension and focus on the interaction between the characters. The setting suggests a private office or meeting room, contributing to the serious and intense atmosphere of the conversation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a suit sits across a table from another man, looking serious as he holds a knife. The scene is intense and dramatic.
5.005005
1,920
810
120
1920x810
short_scene_001_short_12_Angry_Men_(3_10)_Movie_CLIP_-_Who_Changed_Their_Vote?_(1957)_HD
4.793395
{'frame_00:00.png': 5.349264144897461, 'frame_00:01.png': 5.072165489196777, 'frame_00:02.png': 4.439499855041504, 'frame_00:03.png': 4.600992202758789, 'frame_00:04.png': 4.5050554275512695}
short_12_Angry_Men_(3_10)_Movie_CLIP_-_Who_Changed_Their_Vote?_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(3_10)_Movie_CLIP_-_Who_Changed_Their_Vote?_(1957)_HD/short_scene_001.mp4
0.073746
0.81655
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
5
4.6
0
{0: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 1: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 2: {'raw_counts': {'person': 4, 'tie': 1, 'total': 5}, 'bucketed_counts': {'humans': 4, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 5}}, 3: {'raw_counts': {'person': 8, 'total': 8}, 'bucketed_counts': {'humans': 8, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 8}}, 4: {'raw_counts': {'person': 7, 'tie': 1, 'total': 8}, 'bucketed_counts': {'humans': 7, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 8}}}
0.021503
[-1 1 1 -1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=EqDd06GW76o
scene_001
1
1,150
1,270
00:00:47.965
00:00:52.970
47.964631
52.969636
5.005005
47.964631
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with a concerned expression is captured in close-up, his face showing signs of distress and perhaps anxiety. The background is slightly blurred, suggesting the setting is a room with other people present, though they are not the focus. The camera angle highlights his emotional state, emphasizing the lines and contours of his face. The scene transitions, shifting to show the man from the back as he faces a group of other men, some in casual attire, seated or standing around a table. The composition suggests a tense or serious discussion, with the framing indicating a moment of confrontation or deliberation. The lighting is consistent, providing a stark contrast and highlighting the gravity of the situation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds in a somber room, with a man in the foreground displaying a serious expression. As the camera pans, it reveals a group of men seated around a table, engaged in intense discussion and creating an atmosphere charged with tension.
5.005005
1,920
810
120
1920x810
short_scene_001_short_12_Angry_Men_(5_10)_Movie_CLIP_-_Re-enactment_(1957)_HD
5.123106
{'frame_00:00.png': 5.146502494812012, 'frame_00:01.png': 5.175464630126953, 'frame_00:02.png': 5.308682441711426, 'frame_00:03.png': 5.2535810470581055, 'frame_00:04.png': 4.731299877166748}
short_12_Angry_Men_(5_10)_Movie_CLIP_-_Re-enactment_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(5_10)_Movie_CLIP_-_Re-enactment_(1957)_HD/short_scene_001.mp4
-0.019598
0.474586
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.2
1.2
0
{0: {'raw_counts': {'tie': 1, 'person': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 1: {'raw_counts': {'tie': 1, 'person': 2, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 3}}, 2: {'raw_counts': {'tie': 1, 'person': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 3: {'raw_counts': {'tie': 1, 'person': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 4: {'raw_counts': {'tie': 1, 'person': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}}
0.021503
[-1 1 1 -1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=1fsFQ2gF4oE
scene_001
1
330
431
00:00:13.764
00:00:17.976
13.763764
17.97631
4.212546
13.763764
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Captured in a medium shot, a middle-aged man is positioned indoors, facing forward with a slight angle to his right; the lighting emphasizes the creases on his face and his dark hair. He's wearing a light-colored, short-sleeved dress shirt and a dark tie, with sweat visible around his armpits. His expression shifts from a wide-mouthed expression to a more subdued, thoughtful gaze. Behind him, a wardrobe is visible, with several dark coats or jackets hanging inside, adding depth to the indoor setting. The scene is presented in black and white.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with a concerned expression wears a sweaty shirt and tie in a room with garments hanging in the background. He appears to be speaking or reacting to something happening off-screen.
4.212546
1,920
810
101
1920x810
short_scene_001_short_12_Angry_Men_(7_10)_Movie_CLIP_-_Down_&_In_(1957)_HD
4.879718
{'frame_00:00.png': 5.414283275604248, 'frame_00:01.png': 5.455104827880859, 'frame_00:02.png': 5.237608909606934, 'frame_00:03.png': 4.354500770568848, 'frame_00:04.png': 4.446122169494629, 'frame_00:05.png': 4.370686054229736}
short_12_Angry_Men_(7_10)_Movie_CLIP_-_Down_&_In_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(7_10)_Movie_CLIP_-_Down_&_In_(1957)_HD/short_scene_001.mp4
-0.040601
0.521126
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:05.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
8.166667
6.666667
0
{0: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 1: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 2: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 3: {'raw_counts': {'person': 15, 'tie': 1, 'chair': 1, 'total': 17}, 'bucketed_counts': {'humans': 15, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 2, 'total': 17}}, 4: {'raw_counts': {'person': 12, 'tie': 1, 'chair': 2, 'total': 15}, 'bucketed_counts': {'humans': 12, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 3, 'total': 15}}, 5: {'raw_counts': {'person': 10, 'tie': 1, 'chair': 1, 'total': 12}, 'bucketed_counts': {'humans': 10, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 2, 'total': 12}}}
0.019524
[-1 1 1 -1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 -1 1 1 1 1 1]
https://youtube.com/watch?v=U44_sEUJROI
scene_001
1
3,401
3,528
00:02:21.850
00:02:27.147
141.850184
147.147147
5.296964
141.850184
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Captured in close-up, a man in a light-colored shirt and dark tie sits with a furrowed brow, looking intensely at something to the left. His face shifts as his brow furrows further, and a slight grimace appears. The camera angle remains steady, focusing solely on his facial expressions and upper body. The background is dimly lit, suggesting an office or courtroom setting, with the edge of another person's shirt visible on the left. The compilation of images includes additional scenes of men seated around a table in a wood-paneled room, seemingly in deep discussion. There's also a promotional poster with the words "Life is in their hands - Death is on their minds!" against a yellow background. A man’s face is frozen in mid-scream, captured in stark lighting, and lastly a man stares intensely with his mouth slightly agape.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a shirt and tie is seen in a tight close-up, exhibiting a stern expression that transitions to one of disapproval. The scene's intensity is heightened by the use of black and white, which contributes to the overall somber mood.
5.296964
1,920
832
127
1920x832
short_scene_001_short_12_Angry_Men_(8_10)_Movie_CLIP_-_These_People_(1957)_HD
4.807707
{'frame_00:00.png': 4.828805446624756, 'frame_00:01.png': 4.621840476989746, 'frame_00:02.png': 4.643094062805176, 'frame_00:03.png': 4.971719264984131, 'frame_00:04.png': 4.973074436187744}
short_12_Angry_Men_(8_10)_Movie_CLIP_-_These_People_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(8_10)_Movie_CLIP_-_These_People_(1957)_HD/short_scene_001.mp4
-0.166125
0.072027
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.4
1.4
0
{0: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 1: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 2: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 3: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 4: {'raw_counts': {'person': 1, 'tie': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}}
-0.026816
[-1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=TXlHKTPfLVA
scene_001
1
95
215
00:00:03.962
00:00:08.967
3.962296
8.967301
5.005005
3.962296
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a light-colored shirt and tie is shown seated in a room, framed by portions of other individuals on either side of him. The scene is shot in black and white, lending it a classic feel. The man gesticulates and speaks with evident intensity, his facial expressions shifting from stern to contemplative. The background suggests a formal setting, possibly a courtroom or a meeting room with traditional paneling. The lighting accentuates his features and adds depth to the scene. The camera focuses on the man's expressions, capturing moments of frustration, determination, and reflection as he speaks.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a button-down shirt and tie appears to be upset while seated at a table. The black and white shot gives a sense of drama.
5.005005
1,920
810
120
1920x810
short_scene_001_short_12_Angry_Men_(9_10)_Movie_CLIP_-_Nose_Marks_(1957)_HD
5.124554
{'frame_00:00.png': 5.237451553344727, 'frame_00:01.png': 5.036994457244873, 'frame_00:02.png': 5.179314613342285, 'frame_00:03.png': 5.1079559326171875, 'frame_00:04.png': 5.06105375289917}
short_12_Angry_Men_(9_10)_Movie_CLIP_-_Nose_Marks_(1957)_HD
../assets/split_scenes/short_12_Angry_Men_(9_10)_Movie_CLIP_-_Nose_Marks_(1957)_HD/short_scene_001.mp4
0.107718
0.044172
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
4
3
0
{0: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 1: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 2: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 3: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}, 4: {'raw_counts': {'person': 3, 'tie': 1, 'total': 4}, 'bucketed_counts': {'humans': 3, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 4}}}
0.017523
[-1 1 1 -1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=RGF6Qyvz2no
scene_001
1
79
199
00:00:03.295
00:00:08.300
3.294962
8.299967
5.005005
3.294962
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In a black and white film scene, a man in a suit and tie is seated at a table, looking to his left with a slight smile on his face. He appears to be engaged in a conversation with someone who is partially visible, their face out of frame, but their gray hair and the back of their head are visible to the right. The suit-clad man is positioned to the left of the frame, facing towards the right, creating a sense of interaction between the two individuals. The background features dark wooden paneling, adding depth and a formal atmosphere to the setting. The lighting is subdued, emphasizing the contours of the man's face and clothing, and highlighting the textures of the wooden backdrop.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a suit converses with another man in a black-and-white setting. The lighting emphasizes their expressions as they engage in dialogue.
5.005005
1,920
810
120
1920x810
short_scene_001_short_12_Monkeys_(10_10)_Movie_CLIP_-_Zoo_Animals_Run_Free_(1995)_HD
5.004919
{'frame_00:00.png': 4.982006072998047, 'frame_00:01.png': 4.866259574890137, 'frame_00:02.png': 4.870838642120361, 'frame_00:03.png': 5.300572395324707}
short_12_Monkeys_(10_10)_Movie_CLIP_-_Zoo_Animals_Run_Free_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(10_10)_Movie_CLIP_-_Zoo_Animals_Run_Free_(1995)_HD/short_scene_001.mp4
0.016794
0.098775
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}]}
1
0
1
{"0": {"raw_counts": {"dog": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"dog": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"dog": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"dog": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.006777
[ 1 1 1 1 -1 1 1 1 1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=HigxGvmHEdQ
scene_001
1
193
270
00:00:08.050
00:00:11.261
8.049716
11.261261
3.211545
8.049716
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The shot captures a wide view of a cityscape at dusk, framed by the statues of animals in the foreground. Dominating the right side of the frame is a large sculpture of a bear, lying atop what appears to be its prey, positioned on a stone platform. To the left of the bear, another sculpture, possibly a bull or bison, stands on a similar pedestal. The steps leading up to the statues are lightly dusted with snow, hinting at the cold weather. A tiger walks up the snow-covered steps towards the statue of the bear. The buildings in the background vary in height, with a couple of taller structures standing out against the sky. The sky transitions from a soft orange near the horizon to a muted grey as it rises higher, creating a subtle gradient. Bare trees are interspersed among the buildings, adding to the wintery feel of the scene. The light is dim, casting long shadows and highlighting the textures of the stone and the fur of the animal sculptures.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A tiger walks along a snow-covered staircase, with city buildings and sculptures in the background. The scene is cast in soft, overcast lighting, creating a serene atmosphere.
3.211545
1,280
516
77
1280x516
short_scene_001_short_12_Monkeys_(1_10)_Movie_CLIP_-_The_Scientists'_Offer_(1995)_HD
5.076098
{'frame_00:00.png': 5.081973075866699, 'frame_00:01.png': 5.110137939453125, 'frame_00:02.png': 5.030533313751221, 'frame_00:03.png': 5.081747055053711}
short_12_Monkeys_(1_10)_Movie_CLIP_-_The_Scientists'_Offer_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(1_10)_Movie_CLIP_-_The_Scientists'_Offer_(1995)_HD/short_scene_001.mp4
0.055701
0.120119
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
7.25
6.25
0
{0: {'raw_counts': {'person': 6, 'clock': 1, 'total': 7}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 7}}, 1: {'raw_counts': {'person': 7, 'clock': 1, 'total': 8}, 'bucketed_counts': {'humans': 7, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 8}}, 2: {'raw_counts': {'person': 6, 'clock': 1, 'total': 7}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 7}}, 3: {'raw_counts': {'person': 6, 'clock': 1, 'total': 7}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 7}}}
-0.063049
[-1 1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=TpGPSRGrL3s
scene_001
1
121
203
00:00:05.047
00:00:08.467
5.046713
8.4668
3.420087
5.046713
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Seven individuals, clad in white lab coats, occupy a scene filled with complex, antiquated machinery. The set is crammed with gears, clocks, and electronic equipment, creating an atmosphere of elaborate, yet dated, technology. Each person sits at a station, attending to a different piece of equipment, with the arrangement suggesting a coordinated operation. The scene features a central figure operating a typewriter, while the others are engaged with various devices, hinting at a meticulously synchronized process. The overall setting evokes a sense of steampunk science.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people wearing white coats and glasses are stationed at a complex array of mechanical and electrical equipment in a retro-futuristic setting. They appear to be operating the machinery with focused expressions, immersed in the task at hand.
3.420087
1,280
540
82
1280x540
short_scene_001_short_12_Monkeys_(2_10)_Movie_CLIP_-_Cole_Meets_Kathryn_(1995)_HD
4.30509
{'frame_00:00.png': 4.384446144104004, 'frame_00:01.png': 4.297839164733887, 'frame_00:02.png': 4.4099040031433105, 'frame_00:03.png': 4.103271007537842, 'frame_00:04.png': 4.3299880027771}
short_12_Monkeys_(2_10)_Movie_CLIP_-_Cole_Meets_Kathryn_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(2_10)_Movie_CLIP_-_Cole_Meets_Kathryn_(1995)_HD/short_scene_001.mp4
0.270531
1.166939
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.4
2.4
0
{0: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 1: {'raw_counts': {'person': 4, 'total': 4}, 'bucketed_counts': {'humans': 4, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 4}}, 2: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 3: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 4: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}}
0.010046
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=e4yCxKv-2qw
scene_001
1
135
255
00:00:05.631
00:00:10.636
5.630631
10.635636
5.005005
5.630631
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In an enclosed setting, a woman with long, wavy hair sits face-to-face with a bald man who appears to be covered in some kind of substance. The woman is positioned on the right side of the frame, while the man occupies the left, their gazes meeting. They are separated by what appears to be a wire mesh. The woman is wearing a dark, long-sleeved top, and the man's head and shoulders are visible, showing a shiny, textured covering over his skin. The setting is dimly lit, casting shadows that emphasize the characters' expressions and the textures of the surfaces around them. The atmosphere suggests a serious or intense conversation, given their focused expressions and close proximity within the confined space.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with long hair is sitting in a room, talking to a bald man with a fleshy appearance. The scene appears to be a tense or intimate conversation, set against a backdrop of metallic grids.
5.005005
1,280
540
120
1280x540
short_scene_001_short_12_Monkeys_(3_10)_Movie_CLIP_-_Plague_of_Madness_(1995)_HD
4.311426
{'frame_00:00.png': 4.244767189025879, 'frame_00:01.png': 4.307999134063721, 'frame_00:02.png': 4.181949138641357, 'frame_00:03.png': 4.486663818359375, 'frame_00:04.png': 4.335752010345459}
short_12_Monkeys_(3_10)_Movie_CLIP_-_Plague_of_Madness_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(3_10)_Movie_CLIP_-_Plague_of_Madness_(1995)_HD/short_scene_001.mp4
1.277561
2.340511
{'frame_00:00.png': [{'<COLOR>': 'White'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
1.4
1.2
0
{0: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 1: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 2: {'raw_counts': {'person': 1, 'couch': 1, 'total': 2}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 2}}, 3: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 4: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}}
-0.068345
[-1 -1 1 1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=9l_U2xXb9VI
scene_001
1
108
228
00:00:04.505
00:00:09.510
4.504505
9.50951
5.005005
4.504505
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person wearing patterned pajama pants and brown and white shoes sits in a large black recliner chair. The person is initially covered with a tan blanket or robe, and removes it from their head, revealing a tan sweater. The chair is in a bright, institutional-looking room with white walls and tile floors. Another person's legs and feet are briefly visible in the background as they walk out of frame. The person in the chair looks up and to the left.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in patterned pajama pants sits in a black leather recliner, initially covered with brown and white fabrics, which are then removed as the person sits up and looks to the side. Another person with partially visible legs walks in the background.
5.005005
1,280
598
120
1280x598
short_scene_001_short_12_Monkeys_(4_10)_Movie_CLIP_-_Institutionalized_With_Jeffrey_(1995)_HD
4.681987
{'frame_00:00.png': 5.097666263580322, 'frame_00:01.png': 4.411544322967529, 'frame_00:02.png': 4.842087268829346, 'frame_00:03.png': 4.492862224578857, 'frame_00:04.png': 4.56577730178833}
short_12_Monkeys_(4_10)_Movie_CLIP_-_Institutionalized_With_Jeffrey_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(4_10)_Movie_CLIP_-_Institutionalized_With_Jeffrey_(1995)_HD/short_scene_001.mp4
-0.181351
0.759313
{'frame_00:00.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}]}
2.2
2
0.2
{0: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 1: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 2: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 3: {'raw_counts': {'person': 2, 'total': 2}, 'bucketed_counts': {'humans': 2, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 2}}, 4: {'raw_counts': {'person': 2, 'cat': 1, 'total': 3}, 'bucketed_counts': {'humans': 2, 'animals': 1, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 3}}}
0.006417
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=wcztDZ13TLI
scene_001
1
635
734
00:00:26.485
00:00:30.614
26.484818
30.613947
4.129129
26.484818
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men are engaged in a tense conversation, the camera focused on their upper bodies. The man in the foreground has a shaved head with some markings on the back, while the man facing him has short, neatly styled hair and is wearing a collared shirt. The man in the collared shirt gestures with his hands as he speaks, his expression animated. Behind them, a doorway reveals a figure walking away down a long, arched hallway. The setting appears to be inside a facility with white tiled walls.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in a tan sweatshirt stands facing another individual with a shaved head and a tattoo, as they engage in a conversation within a building hallway. A silhouette of a figure can be seen walking by in the background.
4.129129
1,280
684
99
1280x684
short_scene_001_short_12_Monkeys_(5_10)_Movie_CLIP_-_Explaining_to_the_Doctors_(1995)_HD
4.774935
{'frame_00:00.png': 4.81479549407959, 'frame_00:01.png': 4.789970874786377, 'frame_00:02.png': 4.750144004821777, 'frame_00:03.png': 4.7632036209106445, 'frame_00:04.png': 4.756558895111084}
short_12_Monkeys_(5_10)_Movie_CLIP_-_Explaining_to_the_Doctors_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(5_10)_Movie_CLIP_-_Explaining_to_the_Doctors_(1995)_HD/short_scene_001.mp4
0.50188
0.877759
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}]}
11.6
6
0
{0: {'raw_counts': {'person': 6, 'cup': 1, 'dining table': 1, 'cell phone': 1, 'tie': 1, 'total': 10}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 4, 'total': 10}}, 1: {'raw_counts': {'person': 7, 'tie': 3, 'fork': 1, 'cup': 1, 'laptop': 1, 'dining table': 1, 'handbag': 1, 'cell phone': 1, 'total': 16}, 'bucketed_counts': {'humans': 7, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 9, 'total': 16}}, 2: {'raw_counts': {'person': 6, 'tie': 2, 'cup': 1, 'dining table': 1, 'book': 1, 'total': 11}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 5, 'total': 11}}, 3: {'raw_counts': {'person': 6, 'tie': 3, 'dining table': 1, 'chair': 1, 'book': 1, 'total': 12}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 6, 'total': 12}}, 4: {'raw_counts': {'person': 5, 'handbag': 1, 'cell phone': 1, 'book': 1, 'tie': 1, 'total': 9}, 'bucketed_counts': {'humans': 5, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 4, 'total': 9}}}
0.031127
[-1 1 1 1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=q74RKOmIjC8
scene_001
1
98
218
00:00:04.087
00:00:09.092
4.087421
9.092426
5.005005
4.087421
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of individuals, mostly clad in white lab coats, sit around a table in what appears to be a stark, institutional setting, possibly a hospital or research facility, bathed in a hazy, diffused light, where the person in the foreground, facing away from the camera and wearing a white robe, seems to be the focus of the meeting as the others listen attentively. The setting, with its muted color palette and overcast lighting, contributes to a somber, serious atmosphere, while the presence of items such as files, pens, and a tape recorder on the table hints at the professional nature of the gathering. The woman in the background maintains a composed demeanor, subtly suggesting she is an observer or playing a supportive role.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of doctors in white coats sit around a table during a meeting, while a person in a white robe stands at the head of the table, leading the discussion. The scene is dimly lit, creating a serious and professional atmosphere.
5.005005
1,280
638
120
1280x638
short_scene_001_short_12_Monkeys_(6_10)_Movie_CLIP_-_Eating_a_Spider_(1995)_HD
4.750227
{'frame_00:00.png': 4.769765377044678, 'frame_00:01.png': 4.993686676025391, 'frame_00:02.png': 4.964531421661377, 'frame_00:03.png': 4.709323883056641, 'frame_00:04.png': 4.650654315948486, 'frame_00:05.png': 4.536501884460449, 'frame_00:06.png': 4.627125263214111}
short_12_Monkeys_(6_10)_Movie_CLIP_-_Eating_a_Spider_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(6_10)_Movie_CLIP_-_Eating_a_Spider_(1995)_HD/short_scene_001.mp4
0.042887
0.644013
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:05.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:06.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}]}
0.857143
0.714286
0.142857
{0: {'raw_counts': {'total': 0}, 'bucketed_counts': {'humans': 0, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 0}}, 1: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 2: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 3: {'raw_counts': {'dog': 1, 'total': 1}, 'bucketed_counts': {'humans': 0, 'animals': 1, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 4: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 5: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}, 6: {'raw_counts': {'person': 1, 'total': 1}, 'bucketed_counts': {'humans': 1, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 1}}}
-0.021671
[-1 1 1 -1 -1 -1 1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=eb46yXP211w
scene_001
1
73
218
00:00:03.045
00:00:09.092
3.044711
9.092426
6.047714
3.044711
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a hospital gown is seen through a wire mesh, his eyes wide with shock or fear, and a slight bruise on his forehead. The wire mesh creates a barrier, and the man's hands grip it tightly. His mouth is slightly open, expressing surprise or apprehension. The background behind the mesh is dimly lit, suggesting a sterile, institutional environment. The lighting is subdued, casting shadows that emphasize the man's distressed expression.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A bald man with a visible injury looks up with a surprised expression, framed by a grid-like structure suggesting confinement. His face is illuminated by dim, uneven lighting.
6.047714
1,280
540
145
1280x540
short_scene_001_short_12_Monkeys_(7_10)_Movie_CLIP_-_The_Great_Escape_(1995)_HD
5.049436
{'frame_00:00.png': 4.974245548248291, 'frame_00:01.png': 4.87891960144043, 'frame_00:02.png': 5.121516227722168, 'frame_00:03.png': 5.138129234313965, 'frame_00:04.png': 5.134367942810059}
short_12_Monkeys_(7_10)_Movie_CLIP_-_The_Great_Escape_(1995)_HD
../assets/split_scenes/short_12_Monkeys_(7_10)_Movie_CLIP_-_The_Great_Escape_(1995)_HD/short_scene_001.mp4
-0.690408
0.335048
{'frame_00:00.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Blue, White'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Blue, White'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Blue, White'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}]}
5.6
5.2
0
{0: {'raw_counts': {'person': 6, 'total': 6}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 6}}, 1: {'raw_counts': {'person': 6, 'total': 6}, 'bucketed_counts': {'humans': 6, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 6}}, 2: {'raw_counts': {'person': 5, 'total': 5}, 'bucketed_counts': {'humans': 5, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 0, 'total': 5}}, 3: {'raw_counts': {'person': 4, 'sink': 1, 'total': 5}, 'bucketed_counts': {'humans': 4, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 5}}, 4: {'raw_counts': {'person': 5, 'chair': 1, 'total': 6}, 'bucketed_counts': {'humans': 5, 'animals': 0, 'vehicles': 0, 'food': 0, 'others': 1, 'total': 6}}}
-0.058532
[-1 -1 1 1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=cu_A-aG8G7U
scene_001
1
121
241
00:00:05.047
00:00:10.052
5.046713
10.051718
5.005005
5.046713
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene takes place inside a facility and shows a man sitting in a chair with visible injuries on his face. He is wearing a patient gown and appears to be in a state of contemplation or fatigue. Beside him, another man, also in a patient gown, is laughing animatedly. The background reveals other people moving around, some in similar patient attire and others seemingly medical personnel in white coats. A metal-barred window and other institutional elements suggest a hospital or psychiatric facility. The lighting is soft, creating a sterile and slightly unsettling atmosphere. The camera focuses closely on the two men in the foreground, emphasizing the contrast between their emotional states.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person with facial injuries sits in a brightly lit mental institution, while another laughs beside them in a hospital gown, as other patients are visible in the background. The scene shows people in a large, bright room with arched doorways and a high ceiling.
5.005005
1,280
688
120
1280x688
short_scene_001_short_12_Rounds_3__Lockdown_(2015)_-_Firing_Policy_Scene_(2_5)_|_Movieclips
5.119929
{'frame_00:00.png': 5.055117607116699, 'frame_00:01.png': 5.085289478302002, 'frame_00:02.png': 5.2193803787231445}
short_12_Rounds_3__Lockdown_(2015)_-_Firing_Policy_Scene_(2_5)_|_Movieclips
../assets/split_scenes/short_12_Rounds_3__Lockdown_(2015)_-_Firing_Policy_Scene_(2_5)_|_Movieclips/short_scene_001.mp4
-0.010607
0.028867
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1.333333
1
0.333333
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"dog": 1, "person": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.036509
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=8PALGqaFoFI
scene_001
1
99
164
00:00:04.129
00:00:06.840
4.129129
6.840174
2.711044
4.129129
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Captured in a close-up shot, a man with light-colored skin, a mustache, and a stubbly beard, appears to be speaking or listening intently, his eyes wide with a look of concern or surprise; the lighting is subdued, casting subtle shadows across his face, accentuating his facial features as he reacts to something unseen; the background is blurred and dark, suggesting an enclosed or dimly lit space, which further focuses attention on the man's expression and emotional state.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is seen in close-up, expressing concern. He has fair skin, blond hair, and a light beard. The scene is dimly lit, contributing to a serious atmosphere.
2.711044
1,920
898
65
1920x898
short_scene_001_short_12_Rounds_3__Lockdown_(2015)_-_Hard_Wired_Scene_(3_5)_|_Movieclips
4.72463
{'frame_00:00.png': 4.8352742195129395, 'frame_00:01.png': 4.557107925415039, 'frame_00:02.png': 4.7815070152282715}
short_12_Rounds_3__Lockdown_(2015)_-_Hard_Wired_Scene_(3_5)_|_Movieclips
../assets/split_scenes/short_12_Rounds_3__Lockdown_(2015)_-_Hard_Wired_Scene_(3_5)_|_Movieclips/short_scene_001.mp4
0.176264
0.553241
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Yellow, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Yellow, Blue'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}]}
1.666667
1.333333
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "tie": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
-0.016709
[-1 1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=sPlsA3_6hB8
scene_001
1
33
101
00:00:01.376
00:00:04.213
1.376376
4.212546
2.83617
1.376376
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In a dimly lit, smoke-filled environment, a man dressed in a dark shirt is seen holding a phone to his ear. He appears to be engaged in a serious conversation, his expression conveying a sense of urgency and concern. He is positioned near what appears to be a large, blurred, yellow object. A wall with horizontal lines stretches behind him, contributing to the industrial feel of the setting. The atmosphere is heavy, with the smoke adding an element of mystery and tension to the scene.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is seen in a dark and smoky environment talking on a phone, with beams of light cutting through the scene. The scene is dimly lit, giving off an intense and suspenseful mood.
2.83617
1,920
810
68
1920x810
short_scene_001_short_12_Rounds_3__Lockdown_(2015)_-_I_Don't_Play_Nice_Scene_(5_5)_|_Movieclips
4.411905
{'frame_00:00.png': 4.273164749145508, 'frame_00:01.png': 4.811917781829834, 'frame_00:02.png': 4.472927093505859, 'frame_00:03.png': 4.378774642944336, 'frame_00:04.png': 4.310954570770264, 'frame_00:05.png': 4.223693370819092}
short_12_Rounds_3__Lockdown_(2015)_-_I_Don't_Play_Nice_Scene_(5_5)_|_Movieclips
../assets/split_scenes/short_12_Rounds_3__Lockdown_(2015)_-_I_Don't_Play_Nice_Scene_(5_5)_|_Movieclips/short_scene_001.mp4
0.019495
0.114925
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'High contrast, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
1
1
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "5": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.072339
[-1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=ZdhLQ1toP9s
scene_001
1
61
193
00:00:02.544
00:00:08.050
2.544211
8.049716
5.505506
2.544211
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A figure in a dark long-sleeved shirt walks away from the camera, his back facing the viewer, as another individual in the foreground, partially out of focus, holds a gun pointed in his direction, an American flag barely visible behind them. The scene is dimly lit, contributing to an atmosphere of tension, with the focus primarily on the receding figure.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person is seen walking in a hallway, while another figure points a gun in the direction of the subject. The scene conveys a tense, suspenseful atmosphere, with muted lighting adding to the sense of foreboding.
5.505506
1,920
810
132
1920x810
short_scene_001_short_12_Rounds_3__Lockdown_(2015)_-_Rookie_Mistake_Scene_(4_5)_|_Movieclips
4.642742
{'frame_00:00.png': 4.497570991516113, 'frame_00:01.png': 4.547952651977539, 'frame_00:02.png': 4.514873504638672, 'frame_00:03.png': 4.604968547821045, 'frame_00:04.png': 5.048346519470215}
short_12_Rounds_3__Lockdown_(2015)_-_Rookie_Mistake_Scene_(4_5)_|_Movieclips
../assets/split_scenes/short_12_Rounds_3__Lockdown_(2015)_-_Rookie_Mistake_Scene_(4_5)_|_Movieclips/short_scene_001.mp4
-0.323016
0.192604
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Left heavy'}]}
1.8
1.8
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.003065
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 1 1 -1 1 -1 -1 -1 1 1 1 1]
https://youtube.com/watch?v=AQpPxTYahZo
scene_001
1
355
475
00:00:14.806
00:00:19.811
14.806473
19.811478
5.005005
14.806473
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In a dimly lit hallway, two figures emerge through a doorway. The first man is clad in a police uniform, complete with a "POLICE" insignia on the chest. His demeanor is serious. Following closely behind is another man, dressed in casual dark clothing. They move together through the corridor, the scene's lighting casting long shadows and emphasizing the somber mood. The walls around them are made of concrete, with a panel and some form of utility access embedded in one wall.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two figures emerge from a doorway, one wearing a police uniform, with dramatic lighting and an ominous mood. The scene unfolds in a dimly lit, institutional-looking corridor.
5.005005
1,920
810
120
1920x810
short_scene_001_short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Take_Out_the_Technical_Scene_(3_10)_|_Movieclips
4.703685
{'frame_00:00.png': 4.732580661773682, 'frame_00:01.png': 4.76108980178833, 'frame_00:02.png': 4.531744003295898, 'frame_00:03.png': 4.789324760437012}
short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Take_Out_the_Technical_Scene_(3_10)_|_Movieclips
../assets/split_scenes/short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Take_Out_the_Technical_Scene_(3_10)_|_Movieclips/short_scene_001.mp4
-0.504122
0.54046
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Green, Cyan'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Green, Cyan'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Mixed light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Green, Cyan, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Mixed light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Green, Cyan, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Left heavy'}]}
4
2.75
0
{"0": {"raw_counts": {"car": 1, "person": 4, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 5}}, "1": {"raw_counts": {"car": 1, "person": 2, "handbag": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 1, "food": 0, "others": 1, "total": 4}}, "2": {"raw_counts": {"car": 1, "person": 3, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 4}}, "3": {"raw_counts": {"car": 1, "person": 2, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 3}}}
-0.050477
[-1 1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=EL3Ma917bug
scene_001
1
28
121
00:00:01.168
00:00:05.047
1.167835
5.046713
3.878879
1.167835
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds at night, with a cinematic feel. A weathered car sits in the foreground, its windshield covered in droplets, reflecting the ambient light. To the left, a decorative lion statue stands atop a pillar, bathed in a warm, yellow glow. In the background, figures in tactical gear and helmets, armed with rifles, move stealthily down the street, creating a sense of tension and suspense. The buildings line the street creating narrow corridors filled with shadows and unknown dangers.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The image shows armed soldiers moving in a dark, urban setting. The scene is lit with an eerie glow, and a weathered vehicle stands prominently in the foreground, adding to the gritty atmosphere.
3.878879
1,920
810
93
1920x810
short_scene_001_short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_They're_With_Us_Scene_(10_10)_|_Movieclips
4.787926
{'frame_00:00.png': 4.769754409790039, 'frame_00:01.png': 4.704930782318115, 'frame_00:02.png': 4.889092445373535}
short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_They're_With_Us_Scene_(10_10)_|_Movieclips
../assets/split_scenes/short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_They're_With_Us_Scene_(10_10)_|_Movieclips/short_scene_001.mp4
-0.072818
0.249599
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Green, Cyan, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Balanced'}]}
2.333333
0.666667
0
{"0": {"raw_counts": {"person": 1, "chair": 1, "bench": 1, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 3}}, "1": {"raw_counts": {"person": 1, "bench": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"bench": 2, "total": 2}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 2}}}
-0.009466
[-1 1 1 1 -1 1 1 1 1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=OxBd2RQI4eQ
scene_001
1
71
129
00:00:02.961
00:00:05.380
2.961295
5.38038
2.419086
2.961295
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds at dusk, showcasing a rooftop setting overlooking a vast, green field dotted with indistinct buildings and illuminated by scattered lights. A solitary palm tree stands prominently in the center of the field, while smoke drifts gently across the horizon. In the foreground, a series of metallic structures with illuminated panels and wires suggest a technological setup. To the right, satellite dishes add to the sense of a communications or surveillance hub. The overall atmosphere is one of quiet observation against the backdrop of an urban landscape.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene shows an outdoor setting at dusk with a silhouette of a palm tree, buildings and a large field visible in the background. Various pieces of equipment, possibly communication or solar panels, are positioned in the foreground.
2.419086
1,920
774
58
1920x774
short_scene_001_short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Welcome_to_Benghazi_Scene_(1_10)_|_Movieclips
4.389161
{'frame_00:00.png': 4.152843952178955, 'frame_00:01.png': 4.142398834228516, 'frame_00:02.png': 4.87224006652832}
short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Welcome_to_Benghazi_Scene_(1_10)_|_Movieclips
../assets/split_scenes/short_13_Hours__The_Secret_Soldiers_of_Benghazi_(2016)_-_Welcome_to_Benghazi_Scene_(1_10)_|_Movieclips/short_scene_001.mp4
0.018945
0.465812
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Green'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated, Green'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}]}
3.666667
3
0.333333
{"0": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "1": {"raw_counts": {"bird": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"car": 1, "person": 5, "total": 6}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 6}}}
-0.096997
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=hgqJjr7pBa0
scene_001
1
70
138
00:00:02.920
00:00:05.756
2.919586
5.755756
2.83617
2.919586
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A chaotic scene unfolds on a bustling street, filled with distressed people fleeing in apparent disarray. In the foreground, a young boy with dark hair gazes directly towards the camera, partially obscured by motion blur, while a muscular man, donning a green head covering and a tactical vest over a black shirt, urgently dashes through the street. An older man with a white cap and glasses attempts to maneuver through the crowd. The backdrop reveals a war-torn cityscape, evidenced by damaged vehicles, debris-strewn streets, and buildings bearing the scars of conflict. A green filter is applied to the shot giving it a distressed and vintage look.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people runs through a war-torn street, conveying a sense of chaos and urgency. Debris and destruction are visible in the background.
2.83617
1,920
810
68
1920x810
short_scene_001_short_2001__A_Space_Odyssey_(1968)_-_Hal_Reads_Lips_Scene_(2_6)_|_Movieclips
5.413613
{'frame_00:00.png': 5.339019775390625, 'frame_00:01.png': 5.446316242218018, 'frame_00:02.png': 5.351513862609863, 'frame_00:03.png': 5.434962272644043, 'frame_00:04.png': 5.496254920959473}
short_2001__A_Space_Odyssey_(1968)_-_Hal_Reads_Lips_Scene_(2_6)_|_Movieclips
../assets/split_scenes/short_2001__A_Space_Odyssey_(1968)_-_Hal_Reads_Lips_Scene_(2_6)_|_Movieclips/short_scene_001.mp4
0.060265
0.240099
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}]}
3.2
1.2
0
{"0": {"raw_counts": {"train": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"train": 1, "person": 3, "motorcycle": 1, "total": 5}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 5}}, "2": {"raw_counts": {"person": 1, "train": 1, "motorcycle": 1, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"motorcycle": 1, "person": 1, "train": 1, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 3}}, "4": {"raw_counts": {"motorcycle": 2, "train": 1, "person": 1, "total": 4}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 3, "food": 0, "others": 0, "total": 4}}}
-0.031331
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=XDO8OYnmkNY
scene_001
1
123
243
00:00:05.130
00:00:10.135
5.13013
10.135135
5.005005
5.13013
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The interior of a spacecraft cockpit is presented with two men wearing grey uniforms. One man faces forward and to the left, while the other faces forward and to the right, both focused on the controls. The cockpit is dimly lit, with red accents along the top and blue buttons on the control panels. Through a central window, another room with a red robot-like figure can be seen, creating a stark contrast with the dark cockpit. The composition emphasizes the tension and focus of the two men, with dramatic lighting highlighting their profiles.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men wearing gray uniforms navigate a futuristic spaceship, illuminated by the soft glow of the control panels and the red interior lighting. The scene evokes a sense of anticipation and technological immersion.
5.005005
1,920
810
120
1920x810
short_scene_001_short_2001__A_Space_Odyssey_(1968)_-_I'm_Afraid_Scene_(4_6)_|_Movieclips
5.317152
{'frame_00:00.png': 5.238839149475098, 'frame_00:01.png': 5.433579444885254, 'frame_00:02.png': 5.341594219207764, 'frame_00:03.png': 5.380237579345703, 'frame_00:04.png': 5.230224609375, 'frame_00:05.png': 5.278438091278076}
short_2001__A_Space_Odyssey_(1968)_-_I'm_Afraid_Scene_(4_6)_|_Movieclips
../assets/split_scenes/short_2001__A_Space_Odyssey_(1968)_-_I'm_Afraid_Scene_(4_6)_|_Movieclips/short_scene_001.mp4
-0.008818
0.036102
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Fluorescent'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Fluorescent'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:05.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Fluorescent'}, {'<COMPOSITION>': 'Center, Symmetrical'}]}
0
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "4": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "5": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}}
-0.133432
[-1 -1 1 1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1]
https://youtube.com/watch?v=HH37JTBpi2A
scene_001
1
339
476
00:00:14.139
00:00:19.853
14.139139
19.853187
5.714047
14.139139
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A figure in a white suit appears to fall through the middle of an abstract and symmetrical room, bathed in intense red light. The room's architecture consists of a grid pattern on the floor and ceiling, with walls adorned with rectangles that seem to be lit from within. The figure's descent is slow, with limbs outstretched as if suspended in space or water. The overall atmosphere is surreal and unsettling, accentuated by the stark color palette and the sense of isolation conveyed by the figure's solitary fall. The lighting is dramatic, enhancing the textures of the room and the contours of the figure.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A figure suspended in space floats in a red-lit chamber with textured walls, creating a surreal and unsettling atmosphere. The scene is dominated by a monochromatic palette and the interplay of light and shadow, emphasizing the isolation and mystery of the figure's confinement.
5.714047
1,920
810
137
1920x810
short_scene_001_short_2001__A_Space_Odyssey_(1968)_-_I'm_Sorry,_Dave_Scene_(3_6)_|_Movieclips
4.925449
{'frame_00:00.png': 4.911889553070068, 'frame_00:01.png': 4.927077293395996, 'frame_00:02.png': 4.970498085021973, 'frame_00:03.png': 4.903609275817871, 'frame_00:04.png': 4.914168357849121}
short_2001__A_Space_Odyssey_(1968)_-_I'm_Sorry,_Dave_Scene_(3_6)_|_Movieclips
../assets/split_scenes/short_2001__A_Space_Odyssey_(1968)_-_I'm_Sorry,_Dave_Scene_(3_6)_|_Movieclips/short_scene_001.mp4
0.016057
0.131523
{'frame_00:00.png': [{'<COLOR>': 'Cool, Saturated, Green, Blue, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Underlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Saturated, Blue, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Saturated, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Saturated, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Underlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Saturated, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Center, Symmetrical'}]}
0.6
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"tv": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}, "2": {"raw_counts": {"tv": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "4": {"raw_counts": {"tv": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}}
-0.042918
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 1 -1]
https://youtube.com/watch?v=Wy4EfdnMZ5g
scene_001
1
99
197
00:00:04.129
00:00:08.217
4.129129
8.21655
4.087421
4.129129
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The view from the cockpit of a spacecraft displays a stark contrast between the illuminated interior and the vast, dark expanse of space outside. A smaller, spherical module is visible in the distance, its lights reflecting off the surrounding darkness. The cockpit itself is filled with panels and monitors, some displaying technical data and graphics in shades of green, pink, and blue. The lighting inside the cockpit casts soft shadows, highlighting the various controls and displays.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
From the cockpit of a spaceship, another spacecraft is viewed through the window as the crew's instruments are displayed on the screen, suggesting a deep-space journey. The atmosphere is tense and anticipatory as the spaceship prepares for a critical operation.
4.087421
1,920
810
98
1920x810
short_scene_001_short_2001__A_Space_Odyssey_Movie_CLIP_-_A_Conversation_with_HAL_(1968)_HD
5.0163
{'frame_00:00.png': 5.0062336921691895, 'frame_00:01.png': 5.0379486083984375, 'frame_00:02.png': 5.006622314453125, 'frame_00:03.png': 5.014396667480469}
short_2001__A_Space_Odyssey_Movie_CLIP_-_A_Conversation_with_HAL_(1968)_HD
../assets/split_scenes/short_2001__A_Space_Odyssey_Movie_CLIP_-_A_Conversation_with_HAL_(1968)_HD/short_scene_001.mp4
0.003407
0.019866
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Red, Orange, Yellow, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Red, Orange, Yellow, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Red, Orange, Yellow, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Red, Orange, Yellow, Blue'}, {'<LIGHTING>': 'High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}]}
0
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}}
-0.019197
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 1 1 -1 -1 -1 -1 1 -1 1 1 -1 -1 -1]
https://youtube.com/watch?v=r13I-TuDcWI
scene_001
1
137
231
00:00:05.714
00:00:09.635
5.714047
9.634635
3.920587
5.714047
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene is dominated by a static, frontal view of a computer interface, its design both sleek and imposing. A circular lens, encased in a metallic frame, features a deep red hue, creating a sense of ominous calm. At the center of the lens, a bright, concentrated point of light shines intensely. The surrounding surfaces of the device are dark and matte, providing a stark contrast that emphasizes the red lens and its light, thus further highlighting the technological and potentially foreboding nature of the setup.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A computer monitor with a singular, glowing red dot dominates the frame, creating an ominous atmosphere. Reflected lights gleam across its surface, suggesting a sleek, advanced technology.
3.920587
1,280
540
94
1280x540
short_scene_001_short_200_Dollars_-_Paper_Moon_(1_8)_Movie_CLIP_(1973)_HD
4.920638
{'frame_00:00.png': 4.991849899291992, 'frame_00:01.png': 4.782049655914307, 'frame_00:02.png': 4.797402381896973, 'frame_00:03.png': 4.862778186798096, 'frame_00:04.png': 5.077093601226807, 'frame_00:05.png': 4.998974323272705, 'frame_00:06.png': 4.934320449829102}
short_200_Dollars_-_Paper_Moon_(1_8)_Movie_CLIP_(1973)_HD
../assets/split_scenes/short_200_Dollars_-_Paper_Moon_(1_8)_Movie_CLIP_(1973)_HD/short_scene_001.mp4
0.155442
0.122543
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:06.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
2.571429
2.285714
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 2, "bowl": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 2, "bowl": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "5": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "6": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
0.007568
[-1, 1, 1, -1, -1, 1, 1, -1, -1, 1, -1, 1, 1, -1, 1, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, 1]
https://youtube.com/watch?v=VBA_fhQoArA
scene_001
1
128
291
00:00:05.339
00:00:12.137
5.338672
12.137137
6.798465
5.338672
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A black and white shot shows a child in a checked hat seated at a table, his mouth open as if speaking or singing, with a man in the background at another table eating and a vintage-style illuminated machine with decorative panels. The child wears a light-colored shirt and overalls, with a glass on the table in front of him, and is positioned centrally in the frame. To the left, the man sits with his back to the camera, absorbed in his meal. The machine, featuring reflective and patterned panels, stands to the right, adding an element of classic design to the setting. The scene is lit in such a way as to highlight the child's expression and the man's focused demeanor, while the machine provides a backdrop of patterned light.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A boy sits at a table in a diner with a man sitting in the background, both eating a meal, captured in black and white. The boy looks forward with an expressive face, while the jukebox stands silently nearby, with subtle shifts in his facial expression and mouth movements.
6.798465
1,280
562
163
1280x562
short_scene_001_short_21_(2008)_-_Beaten_by_the_Pit_Boss_Scene_(7_10)_|_Movieclips
5.457056
{'frame_00:00.png': 5.269359588623047, 'frame_00:01.png': 5.612674713134766, 'frame_00:02.png': 5.388995170593262, 'frame_00:03.png': 5.557193756103516}
short_21_(2008)_-_Beaten_by_the_Pit_Boss_Scene_(7_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Beaten_by_the_Pit_Boss_Scene_(7_10)_|_Movieclips/short_scene_001.mp4
-1.166933
0.823915
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
5
4
0
{"0": {"raw_counts": {"tie": 1, "person": 5, "total": 6}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 6}}, "1": {"raw_counts": {"person": 4, "tie": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "2": {"raw_counts": {"person": 3, "tie": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "3": {"raw_counts": {"person": 4, "tie": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}}
0.03995
[-1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=f30Twj23WqI
scene_001
1
35
109
00:00:01.460
00:00:04.546
1.459793
4.546213
3.08642
1.459793
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds inside a casino, where a well-dressed man in a dark suit stands overseeing a table with several individuals engaged in a game. To the man's left, a young woman with blonde hair sits attentively, while a young man with dark hair leans forward, his expression suggesting involvement in the game. The lighting in the casino is bright, reflecting off the glossy surfaces of the tables and the ornate ceiling details above, creating an opulent and busy environment. The man in the suit watches the players with a composed demeanor, seemingly monitoring the game's progress, while the young players appear focused on their next move.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene takes place in a casino where a man in a suit is interacting with a group of people at a table. The atmosphere appears tense as the man seems to be addressing the group, possibly discussing a game or situation.
3.08642
1,920
810
74
1920x810
short_scene_001_short_21_(2008)_-_Counting_Cards_Scene_(5_10)_|_Movieclips
4.873465
{'frame_00:00.png': 4.792332649230957, 'frame_00:01.png': 4.744795799255371, 'frame_00:02.png': 5.164666175842285, 'frame_00:03.png': 4.792065143585205}
short_21_(2008)_-_Counting_Cards_Scene_(5_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Counting_Cards_Scene_(5_10)_|_Movieclips/short_scene_001.mp4
0.134367
0.469731
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Soft light, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}]}
2
2
0
{"0": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.014026
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=MlCYBrLOP3A
scene_001
1
143
217
00:00:05.964
00:00:09.051
5.964298
9.050717
3.08642
5.964298
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Captured in close-up, a woman with dark hair, wearing a burgundy top, takes a drink from a clear glass with liquid inside. She winces immediately after drinking, her face conveying displeasure. Her eyebrows furrow and she grimaces, suggesting the drink's unpleasant taste. The background is blurred, with bokeh lights creating a soft, atmospheric backdrop behind her. The camera focuses tightly on her facial expressions, capturing her reaction in detail.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman is shown drinking from a glass, with her face showing a slight look of disgust as she swallows. She is in a busy, brightly lit indoor area.
3.08642
1,920
810
74
1920x810
short_scene_001_short_21_(2008)_-_Double-Con_Scene_(10_10)_|_Movieclips
5.221423
{'frame_00:00.png': 5.1824469566345215, 'frame_00:01.png': 5.2895684242248535, 'frame_00:02.png': 5.1922526359558105}
short_21_(2008)_-_Double-Con_Scene_(10_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Double-Con_Scene_(10_10)_|_Movieclips/short_scene_001.mp4
-0.101189
0.215779
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
2.333333
2
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 2, "car": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.010046
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=pz_yVthhLTE
scene_001
1
128
189
00:00:05.339
00:00:07.883
5.338672
7.882883
2.544211
5.338672
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In an exterior shot bathed in dramatic lighting, a man and woman stand facing forward, seemingly surprised. The man is dressed in a short-sleeved shirt and dark pants, while the woman wears a simple black dress. They are positioned within a long, narrow passageway, possibly an underpass or loading dock area. A "NO PARKING LOADING ZONE" sign is visible to the left, while a taxi cab is parked at the far end of the tunnel, its headlights and taillights adding to the scene's illumination. The walls of the passageway are constructed with horizontal concrete blocks, and overhead lights create a series of bright spots along the ceiling. Reflective puddles on the ground suggest recent rain.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two people are standing outside in a dimly lit tunnel, with a taxi waiting in the background. They both appear anxious as they look in the same direction.
2.544211
1,920
810
61
1920x810
short_scene_001_short_21_(2008)_-_Drunken_Fight_Scene_(6_10)_|_Movieclips
4.327184
{'frame_00:00.png': 4.686028957366943, 'frame_00:01.png': 3.7623467445373535, 'frame_00:02.png': 4.533175945281982}
short_21_(2008)_-_Drunken_Fight_Scene_(6_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Drunken_Fight_Scene_(6_10)_|_Movieclips/short_scene_001.mp4
-0.084688
0.238244
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
1.333333
1.333333
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.029944
[-1 -1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=HC4TZdahtgo
scene_001
1
33
93
00:00:01.376
00:00:03.879
1.376376
3.878879
2.502503
1.376376
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A close-up captures a person with fair skin and medium-length brown hair, who is wearing a white and blue pinstriped shirt, as they rest their face in their hand, their gaze directed slightly downward. The person is speaking, with a slight pucker to their lips and a furrow in their brow, which suggests a thoughtful expression. The lighting is dramatic, casting shadows that emphasize the contours of their face. The background is dark and indistinct, creating a sense of intimacy and focus on the subject. The scene transitions to an image of the same person with their hand closer to their mouth, and they're biting their fingers and looking up with a concerned expression.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A young man with dark hair and a striped shirt rests his face in his hand while seemingly listening intently to someone out of frame. With dramatic lighting and a close-up perspective, the scene conveys a sense of contemplation and engagement.
2.502503
1,920
810
60
1920x810
short_scene_001_short_21_(2008)_-_How_to_Count_Cards_Scene_(3_10)_|_Movieclips
5.068765
{'frame_00:00.png': 5.098289489746094, 'frame_00:01.png': 5.089536666870117, 'frame_00:02.png': 5.104437828063965, 'frame_00:03.png': 5.144060134887695, 'frame_00:04.png': 4.907503128051758}
short_21_(2008)_-_How_to_Count_Cards_Scene_(3_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_How_to_Count_Cards_Scene_(3_10)_|_Movieclips/short_scene_001.mp4
-0.023895
0.11326
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
7.6
6.6
0
{"0": {"raw_counts": {"person": 10, "train": 1, "laptop": 1, "total": 12}, "bucketed_counts": {"humans": 10, "animals": 0, "vehicles": 1, "food": 0, "others": 1, "total": 12}}, "1": {"raw_counts": {"person": 6, "train": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 7}}, "2": {"raw_counts": {"person": 6, "train": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 7}}, "3": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "4": {"raw_counts": {"person": 6, "train": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 7}}}
0.032529
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=aiBm3jckAf4
scene_001
1
110
230
00:00:04.588
00:00:09.593
4.587921
9.592926
5.005005
4.587921
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The dim interior of a classic diner is depicted, where a chef in a white uniform works behind a long counter, illuminated by pendant lamps. A vibrant, colorful arcade game stands out near the counter, adding a splash of light to the muted tones. A group of people are seated in a booth against the wall, their faces partially obscured by the shadows. One person looks towards another individual seated across the table, as if engaged in conversation. The overall mood is intimate and somewhat mysterious, with a mix of artificial and natural light casting long shadows and creating a sense of depth within the scene.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people are gathered in a dimly lit diner, seated in booths and enjoying drinks and food, while a chef works at the counter in the background. An arcade game adds a vibrant splash of color to the subdued setting.
5.005005
1,920
810
120
1920x810
short_scene_001_short_21_(2008)_-_Illegal_Gambling_Scene_(4_10)_|_Movieclips
4.410473
{'frame_00:00.png': 4.353404998779297, 'frame_00:01.png': 4.2215681076049805, 'frame_00:02.png': 4.656444549560547}
short_21_(2008)_-_Illegal_Gambling_Scene_(4_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Illegal_Gambling_Scene_(4_10)_|_Movieclips/short_scene_001.mp4
-0.167082
0.682392
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'High contrast, Silhouette, Backlight'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}]}
1
0.666667
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"chair": 1, "person": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
-0.02704
[-1 1 1 -1 -1 1 1 -1 1 1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=mnrxJ_2sfpU
scene_001
1
127
198
00:00:05.297
00:00:08.258
5.296964
8.258258
2.961295
5.296964
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The dimly lit scene frames two figures inside an Asian-themed establishment, as seen through a window with decorative circular patterns. The individual on the left, partially obscured by the window design, is clad in a red shirt and appears to be engaged in a phone call. The figure on the right, mostly in silhouette, seems to be seated, creating an air of mystery. Inside the restaurant, ambient lighting and indistinct forms of other patrons suggest a bustling yet private setting. The focus remains on the interaction between the two figures, their conversation veiled by the window's visual barrier.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two people are seen through a window with ornate decorations, dining inside a warmly lit restaurant. One person appears to be on the phone, while the other is seated across from them.
2.961295
1,920
810
71
1920x810
short_scene_001_short_21_(2008)_-_Secret_Blackjack_Club_Scene_(2_10)_|_Movieclips
4.999551
{'frame_00:00.png': 4.695870876312256, 'frame_00:01.png': 4.742573261260986, 'frame_00:02.png': 5.170311450958252, 'frame_00:03.png': 5.029668807983398, 'frame_00:04.png': 5.3593292236328125}
short_21_(2008)_-_Secret_Blackjack_Club_Scene_(2_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Secret_Blackjack_Club_Scene_(2_10)_|_Movieclips/short_scene_001.mp4
0.187766
0.18538
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1.4
1.4
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.014563
[-1 1 1 1 -1 -1 1 -1 -1 -1 -1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=AcdZaXYh2f0
scene_001
1
155
263
00:00:06.465
00:00:10.969
6.464798
10.969303
4.504505
6.464798
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A young man with dark hair and a red jacket walks into a dimly lit room. He adjusts his jacket, slinging a dark bag over his shoulder, and glances to the side. Behind him, another young man with similar features enters the room, wearing a dark argyle sweater vest. The first man continues to look ahead, while the second man walks with his head down. Both figures appear somber and introspective in the muted lighting.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two young men walk down a dimly lit hallway, one wearing a red jacket and checkered shirt, and the other wearing a patterned sweater. They appear to be in an institutional setting, possibly a school or dormitory.
4.504505
1,920
810
108
1920x810
short_scene_001_short_21_(2008)_-_Winner,_Winner,_Chicken_Dinner_Scene_(1_10)_|_Movieclips
3.877576
{'frame_00:00.png': 3.7902464866638184, 'frame_00:01.png': 4.300738334655762, 'frame_00:02.png': 4.093969821929932, 'frame_00:03.png': 3.502963066101074, 'frame_00:04.png': 3.6999616622924805}
short_21_(2008)_-_Winner,_Winner,_Chicken_Dinner_Scene_(1_10)_|_Movieclips
../assets/split_scenes/short_21_(2008)_-_Winner,_Winner,_Chicken_Dinner_Scene_(1_10)_|_Movieclips/short_scene_001.mp4
-0.081789
0
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
0.2
0
0.2
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "4": {"raw_counts": {"zebra": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.125886
[-1 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 -1 -1 -1 -1 -1 1 -1]
https://youtube.com/watch?v=ds0q8DVz94I
scene_001
1
55
175
00:00:02.294
00:00:07.299
2.293961
7.298966
5.005005
2.293961
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The video opens with a light, hazy background that fades into a dark horizon. The camera then zooms in to a spinning roulette wheel, with many colored chips around it. The roulette wheel stops on the number $1000, with the camera remaining close.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A wide view of a hazy, indistinct landscape is followed by a rotating collection of stacks of coins. The camera zooms in on one of the coins, revealing the number $1000.
5.005005
1,920
810
120
1920x810
short_scene_001_short_27_Dresses_(1_5)_Movie_CLIP_-_Can't_Say_No_(2008)_HD
5.040397
{'frame_00:00.png': 4.759851932525635, 'frame_00:01.png': 5.169393539428711, 'frame_00:02.png': 5.230504989624023, 'frame_00:03.png': 4.963179111480713, 'frame_00:04.png': 5.079054832458496}
short_27_Dresses_(1_5)_Movie_CLIP_-_Can't_Say_No_(2008)_HD
../assets/split_scenes/short_27_Dresses_(1_5)_Movie_CLIP_-_Can't_Say_No_(2008)_HD/short_scene_001.mp4
0.08851
0.256261
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
4.4
3.8
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 5, "cup": 1, "total": 6}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 6}}, "2": {"raw_counts": {"person": 4, "cup": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "3": {"raw_counts": {"person": 4, "cup": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
0.04393
[-1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=r4K7eNzseCI
scene_001
1
204
324
00:00:08.509
00:00:13.514
8.508509
13.513514
5.005005
8.508509
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two individuals are sitting at a bar, engaged in conversation with soft lighting, as one is partially in shadow with the back of their head facing the camera, while the other, with their hair styled in a low bun, looks down pensively. The composition then shifts, revealing a medium shot of both individuals sitting at the bar, now with fairy lights strung in the background, as they continue their conversation, their expressions suggesting a serious discussion. The man on the left, with dark hair, sits beside a makeshift candleholder. As the scene progresses, both individuals are seen seated side by side at the bar, the man running his fingers across his lips, adding to the tension of the moment.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
At a dimly lit bar, a man and a woman sit at the counter, engaged in an emotional conversation. The woman listens intently, while the man seems troubled.
5.005005
1,920
810
120
1920x810
short_scene_001_short_27_Dresses_(3_5)_Movie_CLIP_-_The_Truth_About_Tess_and_George_(2008)_HD
4.747331
{'frame_00:00.png': 4.559791564941406, 'frame_00:01.png': 4.843277931213379, 'frame_00:02.png': 4.838923454284668}
short_27_Dresses_(3_5)_Movie_CLIP_-_The_Truth_About_Tess_and_George_(2008)_HD
../assets/split_scenes/short_27_Dresses_(3_5)_Movie_CLIP_-_The_Truth_About_Tess_and_George_(2008)_HD/short_scene_001.mp4
0.445861
0.30604
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
18.666667
7.666667
0
{"0": {"raw_counts": {"tie": 2, "person": 7, "cup": 3, "chair": 3, "potted plant": 1, "dining table": 1, "wine glass": 2, "total": 19}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 12, "total": 19}}, "1": {"raw_counts": {"person": 9, "cup": 3, "chair": 2, "potted plant": 1, "tie": 3, "dining table": 1, "wine glass": 1, "total": 20}, "bucketed_counts": {"humans": 9, "animals": 0, "vehicles": 0, "food": 0, "others": 11, "total": 20}}, "2": {"raw_counts": {"person": 7, "cup": 3, "chair": 1, "tie": 2, "potted plant": 1, "wine glass": 2, "dining table": 1, "total": 17}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 10, "total": 17}}}
0.03995
[-1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=ts4gt7f_rek
scene_001
1
76
130
00:00:03.170
00:00:05.422
3.169837
5.422089
2.252252
3.169837
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds in a warmly lit, upscale restaurant where four individuals are seated around a table, engaged in an event. An elderly woman, dressed in a light-colored patterned jacket, sits to the left, while to her right, a younger person in a dark suit and light tie appears. A middle-aged man in a suit is seated to the young person's right. The man on the right wipes his mouth with a napkin and the woman on the left is clapping. The table is adorned with champagne flutes, small dishes, and teacups. Behind them, other diners are visible, along with a waiter tending to the tables. The background is richly decorated with flower arrangements and natural branches, adding to the sophisticated ambiance of the dining setting.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people is seated at a table in a restaurant, enjoying a meal and engaging in conversation, as a waiter attends to other guests in the background. The atmosphere appears formal and celebratory, with elegant floral arrangements and sophisticated attire.
2.252252
1,920
810
54
1920x810
short_scene_001_short_27_Dresses_(4_5)_Movie_CLIP_-_I_Think_You_Deserve_More_(2008)_HD
5.09467
{'frame_00:00.png': 5.043787956237793, 'frame_00:01.png': 4.902954578399658, 'frame_00:02.png': 5.024234294891357, 'frame_00:03.png': 5.219892501831055, 'frame_00:04.png': 5.282482147216797}
short_27_Dresses_(4_5)_Movie_CLIP_-_I_Think_You_Deserve_More_(2008)_HD
../assets/split_scenes/short_27_Dresses_(4_5)_Movie_CLIP_-_I_Think_You_Deserve_More_(2008)_HD/short_scene_001.mp4
-1.14941
0.825186
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}]}
4
2.2
0
{"0": {"raw_counts": {"person": 1, "car": 2, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 3}}, "1": {"raw_counts": {"car": 2, "person": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 4, "car": 1, "bus": 1, "total": 6}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 6}}, "3": {"raw_counts": {"person": 2, "car": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 4}}, "4": {"raw_counts": {"person": 2, "car": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 3}}}
0.045901
[-1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=0-HM2VCdrC0
scene_001
1
93
191
00:00:03.879
00:00:07.966
3.878879
7.9663
4.087421
3.878879
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with shoulder-length blonde hair, wearing a black sleeveless dress and a thin necklace, is seen in a medium shot standing on a city street at night, engaging in a tense conversation with a man whose back is to the camera, capturing only a partial view of his right shoulder and head. The woman's expressions alternate between distress and anger as she looks upwards, gestures with her hand, and argues with the man, with trees and parked cars in the background under dim street lighting.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman in a black dress, standing on a city street at night, argues with a man, her expression conveying distress and frustration. The scene is set against a backdrop of parked cars and illuminated buildings.
4.087421
1,920
810
98
1920x810
short_scene_001_short_27_Dresses_(5_5)_Movie_CLIP_-_Get_Over_Here_(2008)_HD
4.318134
{'frame_00:00.png': 4.208882808685303, 'frame_00:01.png': 4.4043073654174805, 'frame_00:02.png': 4.101958274841309, 'frame_00:03.png': 4.374240398406982, 'frame_00:04.png': 4.28245210647583, 'frame_00:05.png': 4.536960124969482}
short_27_Dresses_(5_5)_Movie_CLIP_-_Get_Over_Here_(2008)_HD
../assets/split_scenes/short_27_Dresses_(5_5)_Movie_CLIP_-_Get_Over_Here_(2008)_HD/short_scene_001.mp4
0.197861
0.048722
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:05.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
8.833333
7.5
0
{"0": {"raw_counts": {"person": 8, "tie": 1, "cell phone": 1, "total": 10}, "bucketed_counts": {"humans": 8, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 10}}, "1": {"raw_counts": {"person": 9, "tie": 2, "total": 11}, "bucketed_counts": {"humans": 9, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 11}}, "2": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "3": {"raw_counts": {"person": 7, "tie": 2, "total": 9}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 9}}, "4": {"raw_counts": {"person": 9, "tie": 2, "total": 11}, "bucketed_counts": {"humans": 9, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 11}}, "5": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}}
-0.070824
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=OI2JPJj1d6s
scene_001
1
89
215
00:00:03.712
00:00:08.967
3.712045
8.967301
5.255255
3.712045
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A crowded formal event unfolds with attendees in tuxedos mingling amid the dark ambiance, punctuated by subtle lighting. A woman in a red dress stands amidst the crowd, while another woman looks on, standing beside the man in a tuxedo. The camera captures a man in a casual shirt and jacket threading through the crowd, with a surprised expression, as a man in a tuxedo looks in the same direction. The scene is filled with various faces and details, highlighting the atmosphere of a large, sophisticated gathering.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A crowded scene unfolds with many people dressed in formal wear, suggesting a gala or formal gathering. The setting is dimly lit, which adds a touch of elegance and mystery.
5.255255
1,920
810
126
1920x810
short_scene_001_short_2_Days_in_the_Valley_(3_8)_Movie_CLIP_-_It's_Crooked_(1996)_HD
4.556855
{'frame_00:00.png': 4.2723894119262695, 'frame_00:01.png': 4.862867832183838, 'frame_00:02.png': 4.535306930541992}
short_2_Days_in_the_Valley_(3_8)_Movie_CLIP_-_It's_Crooked_(1996)_HD
../assets/split_scenes/short_2_Days_in_the_Valley_(3_8)_Movie_CLIP_-_It's_Crooked_(1996)_HD/short_scene_001.mp4
1.193975
0.130375
{'frame_00:00.png': [{'<COLOR>': 'Green, White'}, {'<LIGHTING>': 'Hard light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Green, White'}, {'<LIGHTING>': 'Hard light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Green, White'}, {'<LIGHTING>': 'Hard light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}]}
2
1
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "tie": 2, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 3}}, "2": {"raw_counts": {"person": 1, "potted plant": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
-0.019743
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=KGKqdRDo-N8
scene_001
1
34
93
00:00:01.418
00:00:03.879
1.418085
3.878879
2.460794
1.418085
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is captured in a medium shot, his face contorted with either exertion or strong emotion. He is wearing a white shirt, partially unbuttoned to reveal a colorful t-shirt underneath, and a black vest. The background shows a clear, bright sky and a cluster of pine trees on the right, suggesting an outdoor setting. The man's expression and slightly disheveled appearance suggest he is engaged in some intense activity, possibly running or reacting to something alarming.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is seen running outside in the daytime with an expression of panic or urgency. The scene is shot outdoors with a clear sky in the background.
2.460794
1,280
558
59
1280x558
short_scene_001_short_2_Days_in_the_Valley_(4_8)_Movie_CLIP_-_I_Don't_Need_a_Nose_Job?_(1996)_HD
4.366903
{'frame_00:00.png': 4.329619407653809, 'frame_00:01.png': 4.397881031036377, 'frame_00:02.png': 4.373207092285156}
short_2_Days_in_the_Valley_(4_8)_Movie_CLIP_-_I_Don't_Need_a_Nose_Job?_(1996)_HD
../assets/split_scenes/short_2_Days_in_the_Valley_(4_8)_Movie_CLIP_-_I_Don't_Need_a_Nose_Job?_(1996)_HD/short_scene_001.mp4
-0.01872
0.23041
{'frame_00:00.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
3
1.666667
0
{"0": {"raw_counts": {"person": 1, "oven": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"person": 2, "oven": 1, "cell phone": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "2": {"raw_counts": {"person": 2, "oven": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}}
-0.031331
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=OXcI5qu76jY
scene_001
1
63
117
00:00:02.628
00:00:04.880
2.627628
4.87988
2.252252
2.627628
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man stands in a kitchen, stirring a pot on the stove with a wooden spoon in his right hand. In his left hand, he holds a black handgun, pointed downwards. He wears a white t-shirt with a silver chain, a black and beige short-sleeved shirt, and a wristwatch. Beside him is a stainless steel oven and to the left, two decorative globes sit on top of a white refrigerator. Behind him to the right, there is a large painting, depicting a close-up of a woman's torso. The man turns his head to the left, raises his gun and is captured in close-up with an angry expression.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in a black and white shirt stands in a kitchen, holding a gun and stirring a pot on the stove. A large painting of human figures hangs on the wall in the background.
2.252252
1,280
526
54
1280x526
short_scene_001_short_2_Days_in_the_Valley_(5_8)_Movie_CLIP_-_We_Are_the_Police_(1996)_HD
4.32816
{'frame_00:00.png': 4.325369358062744, 'frame_00:01.png': 4.092592239379883, 'frame_00:02.png': 4.291689872741699, 'frame_00:03.png': 4.198762893676758, 'frame_00:04.png': 4.732388019561768}
short_2_Days_in_the_Valley_(5_8)_Movie_CLIP_-_We_Are_the_Police_(1996)_HD
../assets/split_scenes/short_2_Days_in_the_Valley_(5_8)_Movie_CLIP_-_We_Are_the_Police_(1996)_HD/short_scene_001.mp4
0.290358
0.596226
{'frame_00:00.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Green, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Saturated, Green, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Green, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Green, Blue'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}]}
1.6
1.4
0.2
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"dog": 1, "person": 2, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
0.036509
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=mcWrZOrafnA
scene_001
1
78
198
00:00:03.253
00:00:08.258
3.253253
8.258258
5.005005
3.253253
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with long hair and a beard is driving a car on a winding road surrounded by trees. The camera focuses on his face as he speaks, revealing a range of emotions from frustration to contemplation. His expression shifts as he glances to the side, seemingly reacting to something outside the vehicle. The scene unfolds with a natural, candid feel, capturing the man's expressions as he navigates the road and engages in conversation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with a beard is driving a car, looking concerned and speaking. The scene is set in a car with trees visible outside the window, creating a sense of motion.
5.005005
1,280
462
120
1280x462
short_scene_001_short_2_Days_in_the_Valley_(7_8)_Movie_CLIP_-_Come_Out_From_Behind_That_Tree_(1996)_HD
4.382193
{'frame_00:00.png': 4.305116176605225, 'frame_00:01.png': 4.17812967300415, 'frame_00:02.png': 4.3674726486206055, 'frame_00:03.png': 4.468270778656006, 'frame_00:04.png': 4.591973781585693}
short_2_Days_in_the_Valley_(7_8)_Movie_CLIP_-_Come_Out_From_Behind_That_Tree_(1996)_HD
../assets/split_scenes/short_2_Days_in_the_Valley_(7_8)_Movie_CLIP_-_Come_Out_From_Behind_That_Tree_(1996)_HD/short_scene_001.mp4
1.128686
2.751794
{'frame_00:00.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Hard light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Hard light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Hard light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Hard light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Green'}, {'<LIGHTING>': 'Hard light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}]}
2.2
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"car": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"car": 2, "total": 2}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 2, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"car": 3, "truck": 1, "total": 4}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 4, "food": 0, "others": 0, "total": 4}}, "4": {"raw_counts": {"car": 2, "truck": 2, "total": 4}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 4, "food": 0, "others": 0, "total": 4}}}
0.033975
[-1 1 1 1 1 1 1 -1 1 1 -1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=p_UeWtpIW08
scene_001
1
627
747
00:00:26.151
00:00:31.156
26.151151
31.156156
5.005005
26.151151
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A daytime scene unfolds on a quiet residential street, lined with lush trees casting shadows on the pavement. A chain-link fence runs parallel to the road, separating the street from a verdant, grassy area dotted with more trees. As the perspective shifts, several vehicles come into view, parked along the roadside. The scene maintains a serene and suburban atmosphere.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The camera captures a quiet residential street lined with trees, with houses set back behind lawns and fences. As the perspective shifts, several cars come into view, parked along the side of the road under the dappled sunlight filtering through the leaves.
5.005005
1,280
540
120
1280x540
short_scene_001_short_300_(2006)_-_Divine_Power_Scene_(4_5)_|_Movieclips
4.846045
{'frame_00:00.png': 4.824295520782471, 'frame_00:01.png': 4.836131572723389, 'frame_00:02.png': 4.905257701873779, 'frame_00:03.png': 4.820971965789795, 'frame_00:04.png': 4.843566417694092}
short_300_(2006)_-_Divine_Power_Scene_(4_5)_|_Movieclips
../assets/split_scenes/short_300_(2006)_-_Divine_Power_Scene_(4_5)_|_Movieclips/short_scene_001.mp4
-0.270898
0.468343
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center, Symmetrical'}]}
3.2
2.6
0.6
{"0": {"raw_counts": {"cow": 2, "person": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 2, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "1": {"raw_counts": {"person": 2, "cow": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "4": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
-0.046377
[-1 1 1 -1 -1 1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=kmgRv2V_7P4
scene_001
1
119
238
00:00:04.963
00:00:09.927
4.963297
9.926593
4.963297
4.963297
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A large, ornate golden palanquin is carried by a group of people across a rugged, mountainous landscape under a cloudy sky. At the top of the palanquin's steps stands a tall, thin figure. The palanquin is adorned with statues of lions and horned animals, adding to its imposing appearance. A lone figure in a red cape stands at the bottom-left corner of the frame, gazing up at the palanquin with a mix of awe and trepidation, as it moves slowly forward, dominating the desolate environment.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A regal figure stands atop an elaborate, elevated platform carried by a large group of people, moving through a rocky, mountainous landscape. The scene is marked by dramatic lighting and a sense of foreboding.
4.963297
1,280
516
119
1280x516
short_scene_001_short_300_(2006)_-_Remember_Us_Scene_(5_5)_|_Movieclips
5.094415
{'frame_00:00.png': 4.975671291351318, 'frame_00:01.png': 5.139105796813965, 'frame_00:02.png': 5.205145359039307, 'frame_00:03.png': 4.995429515838623, 'frame_00:04.png': 5.156721591949463}
short_300_(2006)_-_Remember_Us_Scene_(5_5)_|_Movieclips
../assets/split_scenes/short_300_(2006)_-_Remember_Us_Scene_(5_5)_|_Movieclips/short_scene_001.mp4
0.133876
0.482938
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}]}
3.2
2.2
0
{"0": {"raw_counts": {"person": 2, "backpack": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "1": {"raw_counts": {"person": 2, "backpack": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "2": {"raw_counts": {"person": 2, "backpack": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "3": {"raw_counts": {"person": 3, "backpack": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.002508
[-1 1 1 1 -1 1 1 1 1 1 -1 1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=nHByIEUb37Y
scene_001
1
193
313
00:00:08.050
00:00:13.055
8.049716
13.054721
5.005005
8.049716
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A figure stands in a golden field of wheat, a spear held in one hand and a shield in the other, with what appears to be a bandage over the eyes. The figure is dressed in a cape and minimal clothing and has a muscular build. The lighting is dramatic, with the sun setting behind the figure, and mountains visible in the background. Dark, ominous clouds fill the sky. In the distance, a figure appears to be running. The scene is filled with a sense of tension and anticipation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A warrior with a head wound is seen walking across a golden field with his spear and shield, with a second figure shown in the distance. The scene features dramatic lighting and dark clouds in the sky.
5.005005
1,280
540
120
1280x540
short_scene_001_short_300_(2006)_-_This_Is_Sparta!_Scene_(1_5)_|_Movieclips
4.826733
{'frame_00:00.png': 5.049917697906494, 'frame_00:01.png': 4.577976703643799, 'frame_00:02.png': 4.8951029777526855, 'frame_00:03.png': 4.817695617675781, 'frame_00:04.png': 4.792971134185791}
short_300_(2006)_-_This_Is_Sparta!_Scene_(1_5)_|_Movieclips
../assets/split_scenes/short_300_(2006)_-_This_Is_Sparta!_Scene_(1_5)_|_Movieclips/short_scene_001.mp4
0.881053
2.094496
{'frame_00:00.png': [{'<COLOR>': 'Warm, White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, White'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}]}
5.2
5.2
0
{"0": {"raw_counts": {"person": 8, "total": 8}, "bucketed_counts": {"humans": 8, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 8}}, "1": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "4": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}}
0.014026
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=4Prc1UfuokY
scene_001
1
180
282
00:00:07.508
00:00:11.762
7.507508
11.761762
4.254254
7.507508
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds with a group of muscular, bearded men, many draped in cloaks or simple tunics, standing amidst a backdrop of stone columns suggesting an ancient structure. The primary subject, a man with a prominent beard and muscular build, is at the forefront, engaging in intense conversation with someone just outside the frame, his expression serious and focused. Surrounding him are other figures, some gazing intently, others appearing contemplative. The atmosphere is charged with tension and anticipation, hinting at a pivotal moment or confrontation within the narrative. The scene suggests a gathering of warriors or leaders, deliberating on matters of great importance, captured with a cinematic eye that emphasizes the gravity of the situation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of muscular men are seen conversing in an ancient setting, surrounded by stone columns and other figures. The scene exudes a serious and confrontational mood, enhanced by the focused expressions of the men.
4.254254
1,280
540
102
1280x540
short_scene_001_short_300_(2006)_-_This_Is_Where_We_Fight_Scene_(2_5)_|_Movieclips
4.924236
{'frame_00:00.png': 4.7732720375061035, 'frame_00:01.png': 5.056448936462402, 'frame_00:02.png': 4.942988395690918}
short_300_(2006)_-_This_Is_Where_We_Fight_Scene_(2_5)_|_Movieclips
../assets/split_scenes/short_300_(2006)_-_This_Is_Where_We_Fight_Scene_(2_5)_|_Movieclips/short_scene_001.mp4
-1.382468
0.503942
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, White'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}]}
1.666667
1.333333
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "skateboard": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.01428
[-1 1 1 -1 -1 1 1 -1 -1 1 -1 1 1 -1 1 1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=cDpI6Zzy-vo
scene_001
1
202
269
00:00:08.425
00:00:11.220
8.425092
11.219553
2.794461
8.425092
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A warrior in a helmet turns his head to the side, his eyes wide with determination as he prepares for battle, holding a spear firmly in his hand, with a large army visible in the distance, against a backdrop of dramatic, hazy lighting.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A warrior clad in armor turns to face the camera with an intense, determined expression, framed by a vast expanse of what appears to be an ancient battlefield. The scene is lit with dramatic lighting, enhancing the sense of urgency and impending conflict.
2.794461
1,280
540
67
1280x540
short_scene_001_short_300__Rise_of_an_Empire_(2014)_-_Spartan_Rescue_Scene_(10_10)_|_Movieclips
5.159782
{'frame_00:00.png': 5.340054988861084, 'frame_00:01.png': 5.350725173950195, 'frame_00:02.png': 4.939764022827148, 'frame_00:03.png': 5.008583068847656}
short_300__Rise_of_an_Empire_(2014)_-_Spartan_Rescue_Scene_(10_10)_|_Movieclips
../assets/split_scenes/short_300__Rise_of_an_Empire_(2014)_-_Spartan_Rescue_Scene_(10_10)_|_Movieclips/short_scene_001.mp4
0.087205
0.524205
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Yellow'}, {'<LIGHTING>': 'Hard light, High contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}]}
3
3
0
{"0": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "1": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.021872
[-1 -1 1 1 -1 1 1 1 1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=-jtzzs0_bM4
scene_001
1
51
130
00:00:02.127
00:00:05.422
2.127127
5.422089
3.294962
2.127127
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A scene unfolds with two figures in close proximity, poised for conflict. On the left, a figure with dark hair, partially obscured from view, wields a golden sword with ornate detailing on its hilt and wrist guard. The sword is directed towards the right, where a muscular figure stands, bearing the marks of battle on their face and torso. A blue cape drapes over their shoulder, and a sword is pressed against their chest, held by the figure on the left. The background reveals a turbulent body of water, dotted with ships under a cloudy, dramatic sky illuminated by a setting sun. The atmosphere is tense, suggesting a pivotal moment of confrontation.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A dramatic scene shows a man and a woman engaged in a tense standoff, swords clashing with the sea and ships visible in the background. The lighting is dramatic, enhancing the intensity of the moment.
3.294962
1,920
774
79
1920x774
short_scene_001_short_300__Rise_of_an_Empire_(2014)_-_Surrender_to_Me_Scene_(9_10)_|_Movieclips
5.749159
{'frame_00:00.png': 5.798855781555176, 'frame_00:01.png': 5.680508613586426, 'frame_00:02.png': 5.768112659454346}
short_300__Rise_of_an_Empire_(2014)_-_Surrender_to_Me_Scene_(9_10)_|_Movieclips
../assets/split_scenes/short_300__Rise_of_an_Empire_(2014)_-_Surrender_to_Me_Scene_(9_10)_|_Movieclips/short_scene_001.mp4
0.73083
0.065719
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated, Yellow'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated, Yellow'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated, Yellow'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}]}
1.666667
1
0
{"0": {"raw_counts": {"person": 1, "tie": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "tie": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
0.070065
[ 1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 -1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=VOk7mIZRPzI
scene_001
1
1,260
1,327
00:00:52.553
00:00:55.347
52.552553
55.347014
2.794461
52.552553
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with long, dark hair and dramatic eye makeup gazes downward with a serious expression. She is adorned with a dark outfit featuring metallic accents on the shoulders and a unique golden necklace. The background features a somewhat blurred and cloudy sky, with light peeking through the clouds. The overall lighting accentuates her face and the details of her attire.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with long dark hair stares downward with a slight frown while wearing dark clothing with golden decorative elements, against a cloudy sky backdrop. The scene is atmospheric, with a desaturated color palette emphasizing the subject's intense gaze.
2.794461
1,920
790
67
1920x790
short_scene_001_short_300__Rise_of_an_Empire_(2014)_-_The_Ecstasy_Scene_(6_10)_|_Movieclips
5.619147
{'frame_00:00.png': 5.70490837097168, 'frame_00:01.png': 5.555078983306885, 'frame_00:02.png': 5.571148872375488, 'frame_00:03.png': 5.6470513343811035, 'frame_00:04.png': 5.6175456047058105}
short_300__Rise_of_an_Empire_(2014)_-_The_Ecstasy_Scene_(6_10)_|_Movieclips
../assets/split_scenes/short_300__Rise_of_an_Empire_(2014)_-_The_Ecstasy_Scene_(6_10)_|_Movieclips/short_scene_001.mp4
1.259459
0.085733
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Green'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Green'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Green'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Green'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Saturated, Green'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
2
2
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.067376
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 -1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=XDnXI6KXoeg
scene_001
1
469
582
00:00:19.561
00:00:24.274
19.561228
24.274274
4.713046
19.561228
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with short brown hair and a beard faces a dark-haired person whose face is mostly out of frame. The man is wearing a green tunic and looks intently at the other person with a serious expression. The background is dark and appears to be inside a stone structure, with dim light highlighting the man's face and upper body. The scene has a dramatic atmosphere due to the contrast between light and shadow.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with a beard in a green garment glances at another figure beside him. The scene is dimly lit, creating a dramatic and intense mood.
4.713046
1,920
810
113
1920x810
short_scene_001_short_30_Minutes_or_Less_(2011)_-_Hitman_Attack_Scene_(7_9)_|_Movieclips
5.01536
{'frame_00:00.png': 5.008124351501465, 'frame_00:01.png': 4.812470436096191, 'frame_00:02.png': 5.011195659637451, 'frame_00:03.png': 5.229647636413574}
short_30_Minutes_or_Less_(2011)_-_Hitman_Attack_Scene_(7_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_Hitman_Attack_Scene_(7_9)_|_Movieclips/short_scene_001.mp4
-0.079189
0.236451
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.25
1.25
0
{"0": {"raw_counts": {"person": 1, "remote": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"person": 2, "chair": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "2": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "3": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
0.025632
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 1 1 1]
https://youtube.com/watch?v=XC_66AiT5xQ
scene_001
1
125
212
00:00:05.214
00:00:08.842
5.213547
8.842176
3.628629
5.213547
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
From behind, a man with a shaved head raises a pistol, pointing it towards another man, who is older with gray hair, as the older man turns to face him, wearing a gray sweatshirt with "Marines" printed across the front, the scene lit dramatically by the light coming through blinds on the windows behind him.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man holds a gun pointed toward an older man standing in front of a window with blinds. The older man turns to face the gun-holder, wearing a sweatshirt with the word "Marines" across the front.
3.628629
1,920
810
87
1920x810
short_scene_001_short_30_Minutes_or_Less_(2011)_-_I'm_Wearing_a_Bomb_Scene_(2_9)_|_Movieclips
4.904999
{'frame_00:00.png': 4.884653568267822, 'frame_00:01.png': 4.930290222167969, 'frame_00:02.png': 4.8675923347473145, 'frame_00:03.png': 4.622999668121338, 'frame_00:04.png': 5.219459533691406}
short_30_Minutes_or_Less_(2011)_-_I'm_Wearing_a_Bomb_Scene_(2_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_I'm_Wearing_a_Bomb_Scene_(2_9)_|_Movieclips/short_scene_001.mp4
-0.290855
0.272806
{'frame_00:00.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1
1
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
0.013866
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=gQoU4eEkmus
scene_001
1
34
141
00:00:01.418
00:00:05.881
1.418085
5.880881
4.462796
1.418085
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is standing in a hallway with light green walls and a white tiled wall to the right. The man is wearing a plaid shirt, and he has dark hair and dark eyes. A poster of the Declaration of Independence is behind him on the wall to the right. To the left is a bulletin board with colorful papers and a sign that reads "Welcome to..." and has the image of an avocado and a piece of toast. The man's expression goes from a frown to what looks like a crazy, intense expression as he shakes his hands back and forth in front of his face as if indicating an explosion. Then he goes back to a frown.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man stands in a school hallway looking agitated while talking and gesturing with his hands. He appears to be explaining something and looks upset.
4.462796
1,920
810
107
1920x810
short_scene_001_short_30_Minutes_or_Less_(2011)_-_Major_Tan_Scene_(9_9)_|_Movieclips
4.802872
{'frame_00:00.png': 4.843432426452637, 'frame_00:01.png': 4.8904805183410645, 'frame_00:02.png': 4.639766216278076, 'frame_00:03.png': 4.852581977844238, 'frame_00:04.png': 4.788096904754639}
short_30_Minutes_or_Less_(2011)_-_Major_Tan_Scene_(9_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_Major_Tan_Scene_(9_9)_|_Movieclips/short_scene_001.mp4
-0.386787
1.38995
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Magenta'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Saturated, Blue, Magenta'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1.8
1
0
{"0": {"raw_counts": {"person": 1, "tennis racket": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "potted plant": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "4": {"raw_counts": {"person": 1, "potted plant": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
0.041538
[-1 1 1 1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 1 1 -1 -1 1 1 1 1 1]
https://youtube.com/watch?v=ky98ri9AcsU
scene_001
1
149
269
00:00:06.215
00:00:11.220
6.214548
11.219553
5.005005
6.214548
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person is centered in the frame, wearing a light-colored, striped shirt with embroidered designs. The individual has dark hair and a goatee. The background includes a dark screen covered with a camouflage net on the left and leafy green plants on the right. A woven chair is partially visible on the lower left side of the frame. The lighting is even, and the person looks directly at the viewer, with slight head movements throughout the scene. Occasionally, pink text appears on the left side of the frame.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man stands in front of a black backdrop with camouflage netting and foliage, wearing a gold-embroidered shirt. He addresses the camera with a serious expression, as text appears onscreen.
5.005005
1,920
810
120
1920x810
short_scene_001_short_30_Minutes_or_Less_(2011)_-_Rooftop_Confession_Scene_(4_9)_|_Movieclips
4.618946
{'frame_00:00.png': 4.659493923187256, 'frame_00:01.png': 4.578398704528809}
short_30_Minutes_or_Less_(2011)_-_Rooftop_Confession_Scene_(4_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_Rooftop_Confession_Scene_(4_9)_|_Movieclips/short_scene_001.mp4
0.514862
0.538288
{'frame_00:00.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Left heavy'}]}
1.5
1.5
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.006883
[-1 1 1 1 1 1 1 -1 -1 1 -1 -1 1 -1 1 1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=9JkoWs_LIcY
scene_001
1
38
86
00:00:01.585
00:00:03.587
1.584918
3.58692
2.002002
1.584918
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person wearing jeans and a green jacket runs towards the entrance of a brick building, ascending a set of concrete steps to reach the double doors, which feature glass panels and a symbol near the top. The building's facade includes brick columns and a wooden railing, adding architectural detail to the scene. The individual moves with urgency, their body language suggesting haste. The natural light illuminates the scene, highlighting the textures of the brick and concrete.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in a green jacket and blue jeans runs up the steps of a brick building. The person enters the building.
2.002002
1,920
774
48
1920x774
short_scene_001_short_30_Minutes_or_Less_(2011)_-_The_Bank_Robbery_Scene_(5_9)_|_Movieclips
4.651766
{'frame_00:00.png': 4.50413703918457, 'frame_00:01.png': 4.657667636871338, 'frame_00:02.png': 4.793493270874023}
short_30_Minutes_or_Less_(2011)_-_The_Bank_Robbery_Scene_(5_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_The_Bank_Robbery_Scene_(5_9)_|_Movieclips/short_scene_001.mp4
-1.763179
1.117564
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}]}
5.333333
3
0
{"0": {"raw_counts": {"person": 3, "chair": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "1": {"raw_counts": {"person": 3, "chair": 3, "total": 6}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 6}}, "2": {"raw_counts": {"person": 3, "chair": 2, "handbag": 1, "total": 6}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 6}}}
-0.016226
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=v_y3mxio87k
scene_001
1
104
157
00:00:04.338
00:00:06.548
4.337671
6.548215
2.210544
4.337671
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two masked figures burst through a set of glass doors, brandishing handguns towards a uniformed security guard who raises his hands in surrender. The first figure, clad in a green jacket, holds a pistol aloft while the second, wearing a plaid shirt, points his weapon directly forward. A metal detector frame flanks each side of the doorway as the security guard, positioned to the right, faces the intruders with an expression of apprehension. The setting appears to be a building lobby, visible through the glass doors behind the figures.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two masked robbers burst into a building with guns drawn, confronting a security guard with his hands up. The scene unfolds with tension as the robbers advance.
2.210544
1,920
790
53
1920x790
short_scene_001_short_30_Minutes_or_Less_(2011)_-_The_Bomb_Explodes_Scene_(8_9)_|_Movieclips
5.001875
{'frame_00:00.png': 4.939821243286133, 'frame_00:01.png': 5.1100687980651855, 'frame_00:02.png': 5.062098503112793, 'frame_00:03.png': 4.895512104034424}
short_30_Minutes_or_Less_(2011)_-_The_Bomb_Explodes_Scene_(8_9)_|_Movieclips
../assets/split_scenes/short_30_Minutes_or_Less_(2011)_-_The_Bomb_Explodes_Scene_(8_9)_|_Movieclips/short_scene_001.mp4
0.088016
0.207707
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Underlight'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Underlight'}, {'<LIGHTING_TYPE>': 'Moonlight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
3
3
0
{"0": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "1": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
0.010046
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=KqEyt8szupg
scene_001
1
46
133
00:00:01.919
00:00:05.547
1.918585
5.547214
3.628629
1.918585
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of three individuals is positioned inside a car, with the driver focused on the road and the other two passengers displaying expressions of concern or alarm. The driver, on the right, has short hair and wears a red shirt, gripping the steering wheel. The middle passenger, a woman, wears a gray cardigan over a white top and gazes forward with a worried expression. The passenger on the left, a man with a plaid shirt, looks to his right with a similar expression. The scene's lighting suggests it is night, with blurry lights visible through the windshield and the interior dimly lit, amplifying the tension in the enclosed space of the car. In one of the frames, droplets or streaks of water are visible on the windshield, suggesting that there is some sort of precipitation going on outside.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Three passengers sit inside a car, appearing worried as the driver looks ahead. Raindrops hit the windshield as the trio share concerned glances amidst a suspenseful atmosphere.
3.628629
1,920
810
87
1920x810
short_scene_001_short_47_Ronin_(2013)_-_Rescuing_The_Ronin_Scene_(5_10)_|_Movieclips
5.171408
{'frame_00:00.png': 4.9762654304504395, 'frame_00:01.png': 5.326600074768066, 'frame_00:02.png': 5.275981426239014, 'frame_00:03.png': 5.195792198181152, 'frame_00:04.png': 5.199156284332275, 'frame_00:05.png': 5.054654121398926}
short_47_Ronin_(2013)_-_Rescuing_The_Ronin_Scene_(5_10)_|_Movieclips
../assets/split_scenes/short_47_Ronin_(2013)_-_Rescuing_The_Ronin_Scene_(5_10)_|_Movieclips/short_scene_001.mp4
0.712573
1.525434
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}]}
0.833333
0.833333
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "5": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.028507
[-1 1 1 -1 -1 1 -1 -1 -1 -1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=LoBAmFanDhY
scene_001
1
124
251
00:00:05.172
00:00:10.469
5.171839
10.468802
5.296964
5.171839
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A quiet, misty village scene unfolds with traditional Japanese architecture, featuring wooden houses with tiled roofs lining a dirt path. Stone lanterns stand atop small piles of rocks, adding to the serene atmosphere. The fog obscures the background, creating a sense of depth and mystery. Two figures, dressed in period clothing, walk into view from the right, their backs to the camera, as they proceed down the path.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A misty, somber village is depicted, with a few dilapidated Japanese-style structures and stone lanterns. Two figures walk through the scene, evoking a sense of historical drama or suspense.
5.296964
1,920
810
127
1920x810
short_scene_001_short_54_(10_12)_Movie_CLIP_-_Two_Lives_(1998)_HD
4.985294
{'frame_00:00.png': 4.859042167663574, 'frame_00:01.png': 5.076079845428467, 'frame_00:02.png': 5.020760536193848}
short_54_(10_12)_Movie_CLIP_-_Two_Lives_(1998)_HD
../assets/split_scenes/short_54_(10_12)_Movie_CLIP_-_Two_Lives_(1998)_HD/short_scene_001.mp4
0.864556
1.327237
{'frame_00:00.png': [{'<COLOR>': 'Red'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Red'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Firelight'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Red'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}]}
1.333333
1
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "tie": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
0.089922
[ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -1 -1 1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=vzt7Yb_-yiY
scene_001
1
245
298
00:00:10.219
00:00:12.429
10.218552
12.429096
2.210544
10.218552
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman is standing indoors, with a fireplace and Christmas stockings in the background. She is holding a cigarette in her right hand and wearing sunglasses, which she lowers with her left hand. As she lowers her sunglasses, she looks directly at the camera with a slight smile. The light from the fireplace reflects in her eyes. She is wearing a dark turtleneck sweater and small gold hoop earrings. Her dark hair is pulled back from her face.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman is indoors, holding a cigarette in one hand and adjusting dark sunglasses with the other. She briefly removes the glasses to reveal her face before a fireplace decorated with Christmas stockings.
2.210544
1,280
594
53
1280x594
short_scene_001_short_54_(12_12)_Movie_CLIP_-_Steve's_Welcome_Back_Party_(1998)_HD
4.166175
{'frame_00:00.png': 4.108296871185303, 'frame_00:01.png': 3.914400100708008, 'frame_00:02.png': 4.085522651672363, 'frame_00:03.png': 4.289849758148193, 'frame_00:04.png': 4.432807922363281}
short_54_(12_12)_Movie_CLIP_-_Steve's_Welcome_Back_Party_(1998)_HD
../assets/split_scenes/short_54_(12_12)_Movie_CLIP_-_Steve's_Welcome_Back_Party_(1998)_HD/short_scene_001.mp4
0.082688
0.19484
{'frame_00:00.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Saturated, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Mixed light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Saturated, Green, Blue, Purple, Magenta'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light, Mixed light'}, {'<COMPOSITION>': 'Right heavy'}]}
3.8
3.2
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"tie": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 6, "tie": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 7}}, "4": {"raw_counts": {"person": 7, "tie": 1, "total": 8}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 8}}}
-0.030901
[-1 -1 1 -1 -1 1 1 -1 1 1 -1 -1 -1 -1 1 1 1 -1 -1 -1 -1 -1 1 -1 -1 1]
https://youtube.com/watch?v=XTfUTOcdkqY
scene_001
1
98
218
00:00:04.087
00:00:09.092
4.087421
9.092426
5.005005
4.087421
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Inside a crowded club, a man in a patterned shirt is seen up close, partially obscured, with a woman whose face is also partially visible. The scene then shifts to a man in a light-colored suit with a dark shirt and orange tie, surrounded by people, some dancing, under blue and red lights. The lighting is dramatic, casting shadows and highlighting faces in the packed, energetic club environment.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds in a crowded nightclub, where a couple is seen engaged in an intimate embrace, surrounded by a throng of dancing people and dramatic lighting. A man in a suit is later seen standing by himself as he surveys the crowded room.
5.005005
1,280
540
120
1280x540
short_scene_001_short_54_(1_12)_Movie_CLIP_-_Getting_In_(1998)_HD
4.180382
{'frame_00:00.png': 4.098363876342773, 'frame_00:01.png': 4.2418951988220215, 'frame_00:02.png': 4.2008867263793945}
short_54_(1_12)_Movie_CLIP_-_Getting_In_(1998)_HD
../assets/split_scenes/short_54_(1_12)_Movie_CLIP_-_Getting_In_(1998)_HD/short_scene_001.mp4
0.499693
0.571223
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
6
5.666667
0
{"0": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "1": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "2": {"raw_counts": {"person": 7, "tie": 1, "total": 8}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 8}}}
-0.072325
[-1 -1 1 1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=GhSSEuAEcm0
scene_001
1
231
286
00:00:09.635
00:00:11.929
9.634635
11.928595
2.293961
9.634635
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene presents a crowded audience, capturing a range of emotions and expressions. In the foreground, a young woman with short, dark hair and bangs wears a solemn expression, adorned with small, dangling earrings. Adjacent to her, a man in a vibrant, patterned shirt maintains a stoic demeanor. Behind them, a fair-haired individual appears to be cheering or shouting, while another man with long, dark hair and a mustache looks on with a neutral expression. A woman with round glasses stands out to the right, her face exhibiting a mix of excitement and emotion, perhaps on the verge of tears. The composition is tight, emphasizing the collective experience and varied reactions within the crowd, set against a backdrop of diffused lighting that highlights the subjects' faces.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A crowd of people are watching something with anticipation, some smiling and others looking serious, creating a mixed atmosphere of excitement and tension. The diverse group is tightly packed, suggesting they are at an event or performance.
2.293961
1,280
540
55
1280x540
short_scene_001_short_54_(3_12)_Movie_CLIP_-_New_Bus_Boy_(1998)_HD
4.726385
{'frame_00:00.png': 4.707616329193115, 'frame_00:01.png': 4.7946367263793945, 'frame_00:02.png': 4.660542011260986, 'frame_00:03.png': 4.885653495788574, 'frame_00:04.png': 4.583474159240723}
short_54_(3_12)_Movie_CLIP_-_New_Bus_Boy_(1998)_HD
../assets/split_scenes/short_54_(3_12)_Movie_CLIP_-_New_Bus_Boy_(1998)_HD/short_scene_001.mp4
0.338729
0.031777
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Right heavy'}]}
3.8
2.2
0
{"0": {"raw_counts": {"person": 2, "handbag": 1, "bottle": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "1": {"raw_counts": {"person": 2, "handbag": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "2": {"raw_counts": {"person": 2, "handbag": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "3": {"raw_counts": {"person": 2, "handbag": 1, "bottle": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "4": {"raw_counts": {"person": 3, "handbag": 1, "bottle": 1, "total": 5}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 5}}}
0.012017
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=9I0E9w8Nqfg
scene_001
1
120
240
00:00:05.005
00:00:10.010
5.005005
10.01001
5.005005
5.005005
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men are standing indoors. The man on the left is wearing a tan buttoned-up shirt. He has a slight smile and seems to be listening to someone, though his gaze is off-camera to the left. The man on the right is wearing a gray zippered sweatshirt with red accents. He appears to be speaking, with an expression of mild amusement or engagement. The background suggests they are in a medical or laboratory setting, with various containers, bottles, and metal equipment visible. The lighting is cool and subdued, giving the scene a slightly clinical atmosphere.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two young men stand side by side in an indoor environment, one smiling broadly and the other beginning to smile. The scene is dimly lit with neutral-colored surfaces behind them.
5.005005
1,280
540
120
1280x540
short_scene_001_short_54_(5_12)_Movie_CLIP_-_Names_For_the_List_(1998)_HD
4.278361
{'frame_00:00.png': 4.31342077255249, 'frame_00:01.png': 4.098690032958984, 'frame_00:02.png': 4.384766101837158, 'frame_00:03.png': 4.432615280151367, 'frame_00:04.png': 4.162313461303711}
short_54_(5_12)_Movie_CLIP_-_Names_For_the_List_(1998)_HD
../assets/split_scenes/short_54_(5_12)_Movie_CLIP_-_Names_For_the_List_(1998)_HD/short_scene_001.mp4
0.621126
0.556508
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Hard light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
4.4
4
0.2
{"0": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "1": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 3, "dog": 1, "tv": 1, "total": 5}, "bucketed_counts": {"humans": 3, "animals": 1, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
-0.070824
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=Qxn6aafaNP8
scene_001
1
45
165
00:00:01.877
00:00:06.882
1.876877
6.881882
5.005005
1.876877
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene is set in a dimly lit, cluttered room, where a man in a light-colored jacket and dark pants is engaged in conversation with another man wearing a red polo shirt. The man in the jacket gestures with his hands as he speaks, while the man in the polo shirt listens intently. In the foreground, the backs of two individuals with voluminous blonde hair are visible, suggesting they are part of the same group. Behind the men, a wall is adorned with a bulletin board featuring various photos and a "PRESS" sign. To the side, a framed display of record awards and a portrait add to the room's character. The lighting casts shadows, creating an intimate and somewhat mysterious atmosphere within the confined space.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man in a jacket approaches a group of people gathered in an office space, where the lighting is dim and a man in a red shirt stands nearby. The scene features visible window blinds and various items hanging on the wall, creating a busy, lived-in atmosphere.
5.005005
1,280
540
120
1280x540
short_scene_001_short_54_(6_12)_Movie_CLIP_-_Socialites_(1998)_HD
4.840182
{'frame_00:00.png': 5.124741554260254, 'frame_00:01.png': 4.735503196716309, 'frame_00:02.png': 4.803874492645264, 'frame_00:03.png': 4.696609973907471}
short_54_(6_12)_Movie_CLIP_-_Socialites_(1998)_HD
../assets/split_scenes/short_54_(6_12)_Movie_CLIP_-_Socialites_(1998)_HD/short_scene_001.mp4
-0.870293
2.804864
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
3.75
2.75
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 3, "car": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 3, "car": 3, "total": 6}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 3, "food": 0, "others": 0, "total": 6}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
0.04859
[ 1 1 1 1 1 1 1 -1 1 1 1 1 1 -1 1 -1 1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=Y0FAYOIt-VQ
scene_001
1
81
170
00:00:03.378
00:00:07.090
3.378378
7.090424
3.712045
3.378378
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A fair-skinned woman with curly hair and a patterned black and beige long-sleeved blouse is smiling and conversing with a man whose curly hair is visible on the left side of the shot. The woman is holding a martini glass in her left hand and her right hand is gesturing slightly as she speaks. She is wearing a dark green skirt and her face is lit warmly. To the right behind her is a staircase with a red carpet. The lighting on the staircase is relatively dim. The scene takes place in an elegant home. The wall behind the woman is a neutral color.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with curly hair holds a martini glass and smiles as she converses with another person. The scene takes place near a staircase inside a large, well-lit building.
3.712045
1,280
540
89
1280x540
short_scene_001_short_54_(7_12)_Movie_CLIP_-_Disco_Lessons_(1998)_HD
4.391016
{'frame_00:00.png': 4.21326208114624, 'frame_00:01.png': 4.446931838989258, 'frame_00:02.png': 4.628949165344238, 'frame_00:03.png': 4.27492094039917}
short_54_(7_12)_Movie_CLIP_-_Disco_Lessons_(1998)_HD
../assets/split_scenes/short_54_(7_12)_Movie_CLIP_-_Disco_Lessons_(1998)_HD/short_scene_001.mp4
-0.43795
0.645741
{'frame_00:00.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Backlight'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Balanced'}]}
13.5
12.25
0.25
{"0": {"raw_counts": {"person": 14, "backpack": 1, "total": 15}, "bucketed_counts": {"humans": 14, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 15}}, "1": {"raw_counts": {"person": 11, "backpack": 1, "total": 12}, "bucketed_counts": {"humans": 11, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 12}}, "2": {"raw_counts": {"person": 13, "suitcase": 1, "total": 14}, "bucketed_counts": {"humans": 13, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 14}}, "3": {"raw_counts": {"person": 11, "cow": 1, "suitcase": 1, "total": 13}, "bucketed_counts": {"humans": 11, "animals": 1, "vehicles": 0, "food": 0, "others": 1, "total": 13}}}
0.002437
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=uC1Lmk5qK2Q
scene_001
1
304
388
00:00:12.679
00:00:16.183
12.679346
16.18285
3.503504
12.679346
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of young people are in a dance class, moving their hips in unison as they follow the instructor, who stands out in a burgundy dress and high heels; the room is spacious with large windows and a mirrored wall, where the group's reflections are visible; some observers sit in the background, adding to the atmosphere of a learning environment; the overall impression is one of awkwardness and self-consciousness, as the students try to emulate the instructor's moves, with a hula hoop standing idle near the mirrored wall, unused.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people in a dance class practice moving their hips with the instructor in front. The room is bright with sunlight coming through windows.
3.503504
1,280
598
84
1280x598
short_scene_001_short_54_(8_12)_Movie_CLIP_-_Jealousy_(1998)_HD
4.389883
{'frame_00:00.png': 4.40557336807251, 'frame_00:01.png': 4.59238862991333, 'frame_00:02.png': 4.173776149749756, 'frame_00:03.png': 4.362587928771973, 'frame_00:04.png': 4.507044792175293, 'frame_00:05.png': 4.4117350578308105, 'frame_00:06.png': 4.276073932647705}
short_54_(8_12)_Movie_CLIP_-_Jealousy_(1998)_HD
../assets/split_scenes/short_54_(8_12)_Movie_CLIP_-_Jealousy_(1998)_HD/short_scene_001.mp4
-0.194163
0.400124
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Green, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:06.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1
0.857143
0.142857
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "5": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "6": {"raw_counts": {"dog": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
0.048688
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=qPfLNjGXa2M
scene_001
1
218
365
00:00:09.092
00:00:15.224
9.092426
15.223557
6.131131
9.092426
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A young man with blonde curly hair sits indoors, wearing a textured sweater. He looks down with a slightly furrowed brow, then looks up and appears to speak, though his words are not heard. His expression is serious and somewhat concerned. In the background, there is a window with a plant and a candle on the windowsill, along with a turquoise wall and framed artwork. The lighting is soft and diffused.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A young man with curly blond hair sits indoors, wearing a gray sweater and looking down with a serious expression. The scene is softly lit, with a window and a plant with a candle visible in the background.
6.131131
1,280
540
147
1280x540
short_scene_001_short_54_(9_12)_Movie_CLIP_-_The_Clap_(1998)_HD
4.714025
{'frame_00:00.png': 4.880380153656006, 'frame_00:01.png': 4.674185276031494, 'frame_00:02.png': 4.7208099365234375, 'frame_00:03.png': 4.84735107421875, 'frame_00:04.png': 4.447397232055664}
short_54_(9_12)_Movie_CLIP_-_The_Clap_(1998)_HD
../assets/split_scenes/short_54_(9_12)_Movie_CLIP_-_The_Clap_(1998)_HD/short_scene_001.mp4
0.15239
0.521686
{'frame_00:00.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Saturated, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
1.6
1.6
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
0.034188
[-1 1 1 1 -1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=qKerIOG7jdI
scene_001
1
192
312
00:00:08.008
00:00:13.013
8.008008
13.013013
5.005005
8.008008
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with short dark hair and a beaded headband is captured in close-up, her expression shifting from slightly open-mouthed to neutral as she turns her head to the side. The scene is dimly lit with a bluish hue, suggesting a nighttime or indoor setting with dramatic lighting. Behind the woman, blurry figures can be seen, adding depth to the scene and indicating that she is in a crowded environment, possibly at a party or gathering. The focus remains primarily on the woman's face and upper body, highlighting her expression and the details of her headband.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman with dark hair and a headband appears to be speaking intently to someone. The scene is dimly lit with a blueish hue, setting a serious mood.
5.005005
1,280
540
120
1280x540
short_scene_001_short_65_(2023)_-_The_T-Rex_Attack_Scene_|_Movieclips
3.947967
{'frame_00:00.png': 4.097958564758301, 'frame_00:01.png': 3.9189348220825195, 'frame_00:02.png': 4.095281600952148, 'frame_00:03.png': 3.679692268371582}
short_65_(2023)_-_The_T-Rex_Attack_Scene_|_Movieclips
../assets/split_scenes/short_65_(2023)_-_The_T-Rex_Attack_Scene_|_Movieclips/short_scene_001.mp4
0.00004
0.296173
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Moonlight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Underlight'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
0
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}}
-0.138527
[-1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1]
https://youtube.com/watch?v=zisjsD7xQ94
scene_001
1
161
246
00:00:06.715
00:00:10.260
6.715048
10.26026
3.545212
6.715048
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Amidst the dim lighting, a colossal dinosaur head emerges, its features obscured by shadow but its rough texture visible. The creature's head occupies a large portion of the frame, contrasting with a metallic structure on the left, marked by a faint red light. As the scene progresses, the dinosaur's head moves, revealing glimpses of its eye and mouth, showcasing its terrifying presence in the darkened environment.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A large dinosaur thrashes in darkness next to a structure with a red light. The creature's features are obscured by the shadows, adding to the suspenseful mood.
3.545212
1,920
810
85
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Bodice-Ripper_Writer_Scene_|_Movieclips
5.308752
{'frame_00:00.png': 4.532594680786133, 'frame_00:01.png': 5.2031779289245605, 'frame_00:02.png': 5.6543707847595215, 'frame_00:03.png': 5.3651123046875, 'frame_00:04.png': 5.612353324890137, 'frame_00:05.png': 5.484905242919922}
short_80_for_Brady_(2023)_-_Bodice-Ripper_Writer_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Bodice-Ripper_Writer_Scene_|_Movieclips/short_scene_001.mp4
0.026014
0.232872
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Red, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Saturated, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Saturated, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
4.5
4.5
0
{"0": {"raw_counts": {"person": 7, "total": 7}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 7}}, "1": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "3": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "5": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
0.095734
[ 1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 1 1 1 1 1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=nGkh-vVbL8I
scene_001
1
526
658
00:00:21.939
00:00:27.444
21.938605
27.444111
5.505506
21.938605
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds within an indoor setting, likely a locker room, where a fair-skinned man with neatly combed grey hair and a light green jacket is engaged in conversation, his expression shifting between smiling and speaking intently. The camera focuses on him as he speaks, capturing his gestures and facial expressions. The atmosphere is social and convivial, with other people visible in the background, some of whom are engaged in conversation and displaying smiles, all amidst the backdrop of football paraphernalia displayed on the wall behind them, including helmets and jerseys in varying colors. The lighting is bright, illuminating the subjects and the room, contributing to the lively atmosphere of the gathering.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In an indoor setting, a man in a light jacket is clapping and engaged in conversation with another person while surrounded by others. The scene conveys a sense of camaraderie, with a team logo and sports equipment visible in the background.
5.505506
1,920
810
132
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Charity_Poker_Game_Scene_|_Movieclips
4.707856
{'frame_00:00.png': 4.70770263671875, 'frame_00:01.png': 4.929514408111572, 'frame_00:02.png': 4.625049114227295, 'frame_00:03.png': 4.650937080383301, 'frame_00:04.png': 4.626076698303223}
short_80_for_Brady_(2023)_-_Charity_Poker_Game_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Charity_Poker_Game_Scene_|_Movieclips/short_scene_001.mp4
0.12607
0.145578
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
5.8
5.4
0
{"0": {"raw_counts": {"person": 7, "tie": 1, "total": 8}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 8}}, "1": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "2": {"raw_counts": {"person": 5, "tv": 1, "total": 6}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 6}}, "3": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
0.006417
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=iazuONx-0d4
scene_001
1
145
265
00:00:06.048
00:00:11.053
6.047714
11.052719
5.005005
6.047714
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
At a party, a person with light grey hair and wearing a dark, sparkling outfit walks through a crowded hallway. People in cocktail attire hold drinks and talk amongst themselves. Some people stand in the background while others are closer to the camera. A person in a black suit and red shirt looks at their phone. Artwork hangs on the walls, and the lighting is warm and dim. The person in the sparkling outfit exits through a door on the right.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people is gathered in a hallway, seemingly at a party; a woman in a dark, sparkling outfit walks past them toward a doorway. The atmosphere is lively and social as people mingle and converse.
5.005005
1,920
810
120
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Discovering_Tom_Brady_Scene_|_Movieclips
4.580057
{'frame_00:00.png': 4.928205490112305, 'frame_00:01.png': 5.073885440826416, 'frame_00:02.png': 4.9183783531188965, 'frame_00:03.png': 3.591834545135498, 'frame_00:04.png': 4.387982368469238}
short_80_for_Brady_(2023)_-_Discovering_Tom_Brady_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Discovering_Tom_Brady_Scene_|_Movieclips/short_scene_001.mp4
0.367052
0.699642
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}]}
1.6
1.4
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "book": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "3": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}}
-0.039656
[-1 -1 1 1 -1 -1 1 -1 -1 -1 1 1 1 -1 1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 1]
https://youtube.com/watch?v=Ax-nM3rA08Q
scene_001
1
269
389
00:00:11.220
00:00:16.225
11.219553
16.224558
5.005005
11.219553
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene opens with a close-up of an elderly person's hands holding a notepad and a pen. The person is writing something on the notepad, and the words "I didn't know much about football" are visible. The camera zooms out slightly, revealing more of the person's hands and the notepad. The background is blurred, but it appears to be a living room setting. The scene then pulls back, showing the elderly person sitting in a chair, covered with a blanket. A box of tissues sits nearby. The lighting is soft and warm, creating a somber mood.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
An elderly woman is captured in close-up, sitting and writing in a notebook. The camera pulls back to reveal that she is sitting in a chair with a blanket.
5.005005
1,920
810
120
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Doing_Drugs_(on_Accident)_Scene_|_Movieclips
5.242221
{'frame_00:00.png': 5.189666748046875, 'frame_00:01.png': 5.140263080596924, 'frame_00:02.png': 5.412021636962891, 'frame_00:03.png': 5.386911392211914, 'frame_00:04.png': 5.228942394256592, 'frame_00:05.png': 5.199053764343262, 'frame_00:06.png': 5.138687610626221}
short_80_for_Brady_(2023)_-_Doing_Drugs_(on_Accident)_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Doing_Drugs_(on_Accident)_Scene_|_Movieclips/short_scene_001.mp4
0.148449
0.941695
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:05.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:06.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}]}
7.571429
7.571429
0
{"0": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "1": {"raw_counts": {"person": 10, "total": 10}, "bucketed_counts": {"humans": 10, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 10}}, "2": {"raw_counts": {"person": 8, "total": 8}, "bucketed_counts": {"humans": 8, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 8}}, "3": {"raw_counts": {"person": 8, "total": 8}, "bucketed_counts": {"humans": 8, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 8}}, "4": {"raw_counts": {"person": 7, "total": 7}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 7}}, "5": {"raw_counts": {"person": 9, "total": 9}, "bucketed_counts": {"humans": 9, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 9}}, "6": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}}
0.059055
[ 1 1 1 1 1 1 1 1 1 1 1 1 1 -1 1 -1 1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=VeWv_QAxpcg
scene_001
1
135
299
00:00:05.631
00:00:12.471
5.630631
12.470804
6.840174
5.630631
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds in a grand, wood-paneled hall filled with people dressed in formal attire, suggesting a sophisticated gathering. The foreground is framed by lush green plants, creating a sense of depth and enclosure. A woman in a light gray suit stands prominently in the center, engaging with others and holding a champagne glass. To her right, a woman in a light blue dress is conversing with a man in a dark suit. A group of people clusters behind them, their figures partially obscured, adding to the atmosphere of a crowded event. As the scene progresses, other guests enter the frame, including two women in shimmering outfits who join the central group, further enlivening the party atmosphere. The overall impression is one of an elegant, bustling social event, with a warm and inviting ambiance.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Attendees are socializing at a formal event inside a grand hall. The well-dressed people mingle amidst lush greenery, creating an upscale and convivial atmosphere.
6.840174
1,920
810
164
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Fake_Superbowl_Tickets_Scene_|_Movieclips
4.915114
{'frame_00:00.png': 4.856906890869141, 'frame_00:01.png': 5.081393241882324, 'frame_00:02.png': 4.937211990356445, 'frame_00:03.png': 4.837591171264648, 'frame_00:04.png': 4.862467288970947}
short_80_for_Brady_(2023)_-_Fake_Superbowl_Tickets_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Fake_Superbowl_Tickets_Scene_|_Movieclips/short_scene_001.mp4
2.72483
1.786484
{'frame_00:00.png': [{'<COLOR>': 'Red, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Red, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Fluorescent'}, {'<COMPOSITION>': 'Balanced'}]}
10.8
10
0
{"0": {"raw_counts": {"person": 11, "cell phone": 1, "total": 12}, "bucketed_counts": {"humans": 11, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 12}}, "1": {"raw_counts": {"person": 10, "total": 10}, "bucketed_counts": {"humans": 10, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 10}}, "2": {"raw_counts": {"person": 7, "total": 7}, "bucketed_counts": {"humans": 7, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 7}}, "3": {"raw_counts": {"person": 12, "cell phone": 1, "handbag": 2, "total": 15}, "bucketed_counts": {"humans": 12, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 15}}, "4": {"raw_counts": {"person": 10, "total": 10}, "bucketed_counts": {"humans": 10, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 10}}}
0.036502
[-1 1 1 1 1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 1 1 -1 -1 1 1 1 1 1]
https://youtube.com/watch?v=klS-cV4rvNQ
scene_001
1
74
194
00:00:03.086
00:00:08.091
3.08642
8.091425
5.005005
3.08642
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
People are entering a stadium, going through security. There are turnstiles where staff in blue shirts are scanning tickets. Security guards in black and white shirts are monitoring the crowd, some with metal detectors and radios. Many people are wearing team jerseys and carrying bags. The background shows a large banner with sports imagery and the word "Champions." The scene is bustling with activity, people moving through the entrance towards the stadium.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Fans are going through security at an NFL game, including an elderly woman wearing a football jersey and sweatpants. The spectators are entering through turnstiles and metal detectors while security guards supervise.
5.005005
1,920
806
120
1920x806
short_scene_001_short_80_for_Brady_(2023)_-_Inspiring_Tom_Brady_Scene_|_Movieclips
5.012326
{'frame_00:00.png': 4.714048862457275, 'frame_00:01.png': 5.172157287597656, 'frame_00:02.png': 5.150773048400879}
short_80_for_Brady_(2023)_-_Inspiring_Tom_Brady_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Inspiring_Tom_Brady_Scene_|_Movieclips/short_scene_001.mp4
-0.122346
0.237199
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.333333
2
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 3, "umbrella": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}}
0.015534
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 -1 1 1 -1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=p2ud5hQ2ssE
scene_001
1
120
171
00:00:05.005
00:00:07.132
5.005005
7.132132
2.127127
5.005005
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
An eye-level shot shows several individuals in a dimly lit room, likely a command center or control booth. A central figure with short, curly gray hair, wearing a dark-colored sports jersey adorned with the number "12" in glittering silver, is framed by two others wearing headsets. The central person also has a matching sports headband and a string of beads around their neck. These individuals appear to be focused on something out of frame, possibly a sporting event, as they communicate through their headsets. The overall atmosphere suggests tension and concentration, with a focus on teamwork and coordination.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
People are seen in a booth wearing headsets and sporting a team jersey. The atmosphere suggests a tense or high-stakes setting, such as a sports event or competition.
2.127127
1,920
810
51
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_Meeting_Tom_Brady_Scene_|_Movieclips
5.043429
{'frame_00:00.png': 4.645602703094482, 'frame_00:01.png': 4.932485103607178, 'frame_00:02.png': 4.770158767700195, 'frame_00:03.png': 5.4033026695251465, 'frame_00:04.png': 5.465598106384277}
short_80_for_Brady_(2023)_-_Meeting_Tom_Brady_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_Meeting_Tom_Brady_Scene_|_Movieclips/short_scene_001.mp4
-0.607779
0.314889
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Saturated, Red'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
4.6
3.8
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 4, "bottle": 1, "cup": 1, "total": 6}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 6}}, "2": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "3": {"raw_counts": {"person": 4, "cup": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "4": {"raw_counts": {"person": 5, "cup": 1, "total": 6}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 6}}}
0.023974
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 1 1 -1 -1 1 1 1 1 1]
https://youtube.com/watch?v=bkd3RW1iSuU
scene_001
1
88
208
00:00:03.670
00:00:08.675
3.670337
8.675342
5.005005
3.670337
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In a locker room setting, a figure in a sparkling jersey that reads "80 FOR BRADY" on the back approaches a man wearing a gray t-shirt with a Patriots logo. A smiling man in a similar gray shirt is visible in the background. The man in the t-shirt warmly greets the person in the jersey with a hug. Lockers with football helmets and other sports equipment line the background, further establishing the scene's environment.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
In an indoor locker room scene, a person in a jersey walks to a man and embraces him, with other figures looking on. The atmosphere seems warm and supportive.
5.005005
1,920
810
120
1920x810
short_scene_001_short_80_for_Brady_(2023)_-_The_Tom_Brady_Support_Group_Scene_|_Movieclips
4.712677
{'frame_00:00.png': 4.543457508087158, 'frame_00:01.png': 4.73214054107666, 'frame_00:02.png': 4.649622917175293, 'frame_00:03.png': 4.869162082672119, 'frame_00:04.png': 4.7690043449401855}
short_80_for_Brady_(2023)_-_The_Tom_Brady_Support_Group_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_The_Tom_Brady_Support_Group_Scene_|_Movieclips/short_scene_001.mp4
0.100031
0.037399
{'frame_00:00.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light, Fluorescent'}, {'<COMPOSITION>': 'Balanced, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Saturated, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Balanced'}]}
2.8
2
0
{"0": {"raw_counts": {"person": 2, "cup": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 2, "bowl": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "4": {"raw_counts": {"person": 2, "bowl": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}}
0.120642
[ 1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 1 1 1]
https://youtube.com/watch?v=LMaGW5vyASU
scene_001
1
68
188
00:00:02.836
00:00:07.841
2.83617
7.841175
5.005005
2.83617
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two people are seated behind a desk, engaging in a discussion against the backdrop of a "Pats Nation" banner. The individual on the left wears a dark sweatshirt and baseball cap, listening intently with a slight frown, while the person on the right, dressed in a jacket, glasses, and a cap, gestures emphatically with open palms, conveying excitement or explanation. Between them, the desk displays a laptop, football helmets, and a few other sports-related items, creating a cluttered yet themed workspace. The background is filled with sports memorabilia, including shelves laden with trophies, boxing gloves, and various decorative objects, set against a blue wall with names and football imagery, enhancing the sports-centric atmosphere of the scene.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men are sitting at a table surrounded by sports memorabilia, appearing to engage in an animated discussion. The atmosphere is energetic and informal, set against a backdrop of team branding and display shelves.
5.005005
1,920
882
120
1920x882
short_scene_001_short_80_for_Brady_(2023)_-_They're_My_Dancers!_Scene_|_Movieclips
5.342657
{'frame_00:00.png': 5.16606330871582, 'frame_00:01.png': 5.49519157409668, 'frame_00:02.png': 5.383701324462891, 'frame_00:03.png': 5.249217510223389, 'frame_00:04.png': 5.419110298156738}
short_80_for_Brady_(2023)_-_They're_My_Dancers!_Scene_|_Movieclips
../assets/split_scenes/short_80_for_Brady_(2023)_-_They're_My_Dancers!_Scene_|_Movieclips/short_scene_001.mp4
0.869177
1.16304
{'frame_00:00.png': [{'<COLOR>': 'Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Saturated, Red'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}]}
1.8
1.2
0.4
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"person": 1, "dog": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "3": {"raw_counts": {"person": 1, "dog": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "4": {"raw_counts": {"person": 1, "tie": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
0.080473
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 1 1 1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=kMBbmNhOPao
scene_001
1
119
239
00:00:04.963
00:00:09.968
4.963297
9.968302
5.005005
4.963297
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene captures two older women engaged in an intimate conversation at what appears to be a lively outdoor event. The woman on the left is adorned with voluminous blonde hair styled in curls and wearing a navy blue and silver sequined jacket. She gazes with a gentle expression towards the woman on the right. Her companion has short, gray hair and is wearing a headband patterned with stars and stripes, complementing a red and blue top. The background is filled with blurred figures and red clothing, suggesting a gathering of people, likely at a patriotic or celebratory event. The lighting is bright and natural, highlighting the details of their faces and costumes, as they are deeply engrossed in their exchange.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two older women with fair skin are talking closely to one another while dressed in patriotic-themed clothing. The woman with blond hair is gazing at the other woman as she speaks.
5.005005
1,920
882
120
1920x882
short_scene_001_short_8_Mile_(2002)_-_Greg's_Outta_Here_Scene_(7_10)_|_Movieclips
4.937129
{'frame_00:00.png': 5.03442907333374, 'frame_00:01.png': 4.935144424438477, 'frame_00:02.png': 4.901626110076904, 'frame_00:03.png': 4.880531311035156, 'frame_00:04.png': 4.881216049194336, 'frame_00:05.png': 5.055622100830078, 'frame_00:06.png': 4.871331691741943}
short_8_Mile_(2002)_-_Greg's_Outta_Here_Scene_(7_10)_|_Movieclips
../assets/split_scenes/short_8_Mile_(2002)_-_Greg's_Outta_Here_Scene_(7_10)_|_Movieclips/short_scene_001.mp4
-0.239835
0.102619
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:05.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:06.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}]}
1.142857
1
0
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 1, "bus": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 1, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "4": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "5": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "6": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
0.072037
[ 1 1 1 1 1 1 1 -1 -1 1 1 1 1 1 1 -1 -1 1 1 1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=0ShWGyC408I
scene_001
1
107
261
00:00:04.463
00:00:10.886
4.462796
10.885886
6.42309
4.462796
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person is featured in the video's close-up shot inside an interior setting with wood paneling. The person, who has short brown hair, turns to look directly at the camera with a slight smile. The individual then brings a dark-colored cup to their lips and takes a sip, all while maintaining eye contact and a small smirk. The background includes a vintage light fixture, shelves with various items, and kitchen cabinets, contributing to the room's retro feel. The scene is dimly lit, adding to the intimate and somewhat mysterious atmosphere.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man with dark hair drinks from a cup, turns, and smirks at the camera while inside a wood-paneled house. The interior scene is lit by an ornate, old-fashioned lamp.
6.42309
1,280
540
154
1280x540
short_scene_001_short_8_Mile_(2002)_-_Rabbit_Battles_Papa_Doc_Scene_(10_10)_|_Movieclips
4.953808
{'frame_00:00.png': 4.939662933349609, 'frame_00:01.png': 5.154171943664551, 'frame_00:02.png': 5.199042797088623, 'frame_00:03.png': 5.059637546539307, 'frame_00:04.png': 4.4165263175964355}
short_8_Mile_(2002)_-_Rabbit_Battles_Papa_Doc_Scene_(10_10)_|_Movieclips
../assets/split_scenes/short_8_Mile_(2002)_-_Rabbit_Battles_Papa_Doc_Scene_(10_10)_|_Movieclips/short_scene_001.mp4
0.745531
1.236873
{'frame_00:00.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Saturated'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated'}, {'<LIGHTING>': 'Hard light, High contrast, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Cyan, Blue'}, {'<LIGHTING>': 'Hard light, High contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
4.4
3.4
0.8
{"0": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "1": {"raw_counts": {"person": 4, "cow": 1, "dog": 1, "total": 6}, "bucketed_counts": {"humans": 4, "animals": 2, "vehicles": 0, "food": 0, "others": 0, "total": 6}}, "2": {"raw_counts": {"person": 3, "dog": 1, "cow": 1, "total": 5}, "bucketed_counts": {"humans": 3, "animals": 2, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "3": {"raw_counts": {"person": 4, "tie": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 5}}, "4": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}}
-0.009479
[-1 1 1 1 -1 1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=F2hiFbuQ-Qw
scene_001
1
184
304
00:00:07.674
00:00:12.679
7.674341
12.679346
5.005005
7.674341
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene presents a tense face-off between two men, illuminated with dramatic lighting that emphasizes the tension. A third man stands between them, holding a microphone, seemingly ready to announce or mediate. The man on the left shows signs of a recent fight, with a visible bruise around his eye. All three men are captured in close-up, intensifying the focus on their expressions and the immediate atmosphere of confrontation. The background is blurred, indicating a crowd, but the primary focus remains on the central trio and their impending interaction.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men face each other with stern expressions, while a third man stands between them holding a microphone, setting a tense atmosphere. The lighting enhances the drama of the confrontation as the men engage in a face-off.
5.005005
1,280
540
120
1280x540
short_scene_001_short_8_Mile_(2002)_-_The_Lunch_Truck_Scene_(6_10)_|_Movieclips
4.380534
{'frame_00:00.png': 4.331303119659424, 'frame_00:01.png': 3.8675198554992676, 'frame_00:02.png': 3.979708671569824, 'frame_00:03.png': 4.011867046356201, 'frame_00:04.png': 4.536562919616699, 'frame_00:05.png': 4.90809440612793, 'frame_00:06.png': 5.028684139251709}
short_8_Mile_(2002)_-_The_Lunch_Truck_Scene_(6_10)_|_Movieclips
../assets/split_scenes/short_8_Mile_(2002)_-_The_Lunch_Truck_Scene_(6_10)_|_Movieclips/short_scene_001.mp4
0.53643
1.572534
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:05.png': [{'<COLOR>': 'Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:06.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight, Overcast'}, {'<COMPOSITION>': 'Balanced'}]}
4.714286
4.285714
0.428571
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 3, "cow": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 6, "horse": 1, "total": 7}, "bucketed_counts": {"humans": 6, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 7}}, "3": {"raw_counts": {"person": 5, "total": 5}, "bucketed_counts": {"humans": 5, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "5": {"raw_counts": {"person": 4, "cow": 1, "total": 5}, "bucketed_counts": {"humans": 4, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 5}}, "6": {"raw_counts": {"person": 6, "total": 6}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 6}}}
-0.03183
[-1 -1 1 -1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 1 1 -1 1 -1 -1 -1 -1 -1 1 1]
https://youtube.com/watch?v=KQOWuUy-7qE
scene_001
1
94
255
00:00:03.921
00:00:10.636
3.920587
10.635636
6.715048
3.920587
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The camera begins inside a food truck, focusing on the stacked shelves of food containers and food products. As the camera moves away from the truck it reveals the side of a black jacket of a person who appears to be handing out food to a crowd that is gathered on the street outside the truck. The camera shifts to the person handing out the food, focusing on the person's profile, who is wearing a blue bandanna, glasses, and hoop earrings. Behind the person, the frame is filled with a crowd of people, many of whom are holding food containers and some are smoking cigarettes. The scene captures a moment of community and service, set against an urban backdrop.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A group of people stand in a desolate urban setting, with one individual in the foreground wearing a blue bandana and glasses speaking, while others stand around smoking and holding containers. The scene is marked by a sense of somberness and urban decay.
6.715048
1,280
540
161
1280x540
short_scene_001_short_8_Mile_(2002)_-_We're_Being_Evicted_Scene_(4_10)_|_Movieclips
4.591122
{'frame_00:00.png': 4.661941051483154, 'frame_00:01.png': 4.611052513122559, 'frame_00:02.png': 4.500371932983398}
short_8_Mile_(2002)_-_We're_Being_Evicted_Scene_(4_10)_|_Movieclips
../assets/split_scenes/short_8_Mile_(2002)_-_We're_Being_Evicted_Scene_(4_10)_|_Movieclips/short_scene_001.mp4
-1.969868
0.57484
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:01.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}]}
2
1
0
{"0": {"raw_counts": {"person": 1, "tv": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"tv": 1, "person": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "2": {"raw_counts": {"person": 1, "tv": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
-0.054044
[-1 1 1 1 -1 1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 -1 -1 -1 -1 1 1 -1 -1 -1]
https://youtube.com/watch?v=8bY4qPadkSo
scene_001
1
153
215
00:00:06.381
00:00:08.967
6.381381
8.967301
2.585919
6.381381
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene features a vintage television set displaying an animated character resembling a woodpecker. The character is seated in a blue armchair and engrossed in reading a book. The television is situated within a room setting, partially visible are a window with blinds and a person in the background wearing a gray, textured sweater. The room appears to be warmly lit. The focus remains sharp on the TV screen and the animated character.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
An old TV screen displays an animated character sitting in a chair and reading a book while a person in a sweater walks around in the background. The scene has a vintage and nostalgic feel.
2.585919
1,280
540
62
1280x540
short_scene_001_short_8_Mile_(2002)_-_Wink's_Big_Deal_Scene_(2_10)_|_Movieclips
4.633103
{'frame_00:00.png': 4.553243160247803, 'frame_00:01.png': 4.859813690185547, 'frame_00:02.png': 4.511959075927734, 'frame_00:03.png': 4.607394218444824}
short_8_Mile_(2002)_-_Wink's_Big_Deal_Scene_(2_10)_|_Movieclips
../assets/split_scenes/short_8_Mile_(2002)_-_Wink's_Big_Deal_Scene_(2_10)_|_Movieclips/short_scene_001.mp4
0.776984
1.054142
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}]}
2
1
0
{"0": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}, "1": {"raw_counts": {"person": 1, "chair": 2, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 3}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "chair": 1, "total": 2}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 2}}}
0.034501
[-1 1 1 1 1 1 1 -1 -1 1 1 1 1 -1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=LRPEYUlE8rU
scene_001
1
52
129
00:00:02.169
00:00:05.380
2.168836
5.38038
3.211545
2.168836
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A blonde woman stands at a table and sits down. The scene reveals a room that is dimly lit by an overhead lamp and a lamp in the background. The woman sits at the table with her hands on the table surface as she looks ahead. She continues to look ahead and puts her hand on her chin as the camera zooms in to capture more detail. A boombox, CDs, keys, and other items are on the table. A white chair, curtains, and paneled walls are also visible.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A woman sits at a table under a lamp, looking towards the camera, with what appears to be a radio and other items on the table. She appears to be in a dimly lit room.
3.211545
1,280
522
77
1280x522
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Daniel_the_Tiger_Scene_(3_10)_|_Movieclips
5.286742
{'frame_00:00.png': 5.397832870483398, 'frame_00:01.png': 5.154707908630371, 'frame_00:02.png': 5.214561462402344, 'frame_00:03.png': 5.378459453582764, 'frame_00:04.png': 5.28814697265625}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Daniel_the_Tiger_Scene_(3_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Daniel_the_Tiger_Scene_(3_10)_|_Movieclips/short_scene_001.mp4
0.020481
0.338845
{'frame_00:00.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:03.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:04.png': [{'<COLOR>': 'Cool, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Balanced'}]}
3.4
2
0
{"0": {"raw_counts": {"person": 2, "tie": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "1": {"raw_counts": {"person": 2, "tie": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "2": {"raw_counts": {"person": 2, "tie": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "3": {"raw_counts": {"person": 2, "tie": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "4": {"raw_counts": {"person": 2, "tie": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}}
0.021503
[-1 1 1 -1 -1 1 1 -1 -1 1 1 1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=3TKEZqYpx0I
scene_001
1
121
241
00:00:05.047
00:00:10.052
5.046713
10.051718
5.005005
5.046713
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men stand side-by-side within a room bathed in a cool, muted light, creating a somber atmosphere, the man on the left is clad in a gray blazer over a dark turtleneck, his gaze directed to the side, while the man on the right is dressed in a dark suit with a tie and a badge prominently displayed, his expression serious and eyes focused ahead, in the background, there's a glimpse of a child in an orange shirt, adding a touch of innocence to the scene, and a collection of wires and equipment suggests the setting might be a facility or a laboratory.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Two men in suits stand in a dimly lit room, examining something with serious expressions. The atmosphere is tense and mysterious, with a hint of intrigue.
5.005005
1,920
810
120
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Always_Loved_You_Scene_(8_10)_|_Movieclips
5.27607
{'frame_00:00.png': 5.581935882568359, 'frame_00:01.png': 5.259428024291992, 'frame_00:02.png': 5.180625915527344, 'frame_00:03.png': 5.358328342437744, 'frame_00:04.png': 5.000031471252441}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Always_Loved_You_Scene_(8_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Always_Loved_You_Scene_(8_10)_|_Movieclips/short_scene_001.mp4
-0.017147
0.716721
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Desaturated, Blue'}, {'<LIGHTING>': 'Soft light, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Desaturated, Blue'}, {'<LIGHTING>': 'High contrast, Top light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
2.8
2
0.2
{"0": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "1": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "2": {"raw_counts": {"person": 2, "dog": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 3, "baseball glove": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "4": {"raw_counts": {"person": 1, "bed": 1, "baseball glove": 1, "total": 3}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 3}}}
-0.011957
[-1 1 1 -1 -1 1 1 -1 -1 -1 -1 -1 1 -1 1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=QZS_mcnK7mU
scene_001
1
205
325
00:00:08.550
00:00:13.555
8.550217
13.555222
5.005005
8.550217
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene unfolds in a dimly lit room, where a man, partially obscured, is seen holding a glass. His expression is thoughtful, perhaps melancholic, as he gazes into the liquid. A woman with dark hair stands to his left, partially out of focus, contributing to the intimate atmosphere. The lighting is dramatic, casting shadows that add depth and mystery to the moment. The man brings the glass to his lips, possibly savoring a drink, while the woman remains close, their proximity suggesting a close relationship. The scene is carefully framed to draw attention to the man's contemplation and the subtle tension between the two figures.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man sits with a drink in a dimly lit room while looking at old photos, while another person sits out of focus on the left. The scene is somber and contemplative.
5.005005
1,920
810
120
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Don't_Think_You_Are_Broken_Scene_(7_10)_|_Movieclips
5.485022
{'frame_00:00.png': 5.759366512298584, 'frame_00:01.png': 5.480535507202148, 'frame_00:02.png': 5.4119873046875, 'frame_00:03.png': 5.175022125244141, 'frame_00:04.png': 5.598198890686035}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Don't_Think_You_Are_Broken_Scene_(7_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_I_Don't_Think_You_Are_Broken_Scene_(7_10)_|_Movieclips/short_scene_001.mp4
-0.03035
0.83276
{'frame_00:00.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Cyan, Blue'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Side light, Edge light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Cyan'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}]}
3.6
3.6
0
{"0": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "1": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}, "2": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
0.094563
[ 1 1 1 1 1 1 1 -1 1 1 1 1 1 1 1 -1 1 1 1 1 -1 1 1 -1 1 1]
https://youtube.com/watch?v=9AzXX_2BrVk
scene_001
1
321
436
00:00:13.388
00:00:18.185
13.388388
18.184852
4.796463
13.388388
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
Inside a cafe, a man with dark hair sits across from another man, a patterned booth seat behind him, as a woman in a white top and jeans walks by outside the window. The first man looks at the other as he speaks, a slight smile on his face. Another person sits in the background at the left side of the frame, partially out of focus. The scene appears to be in a daylight setting with natural light filtering in from outside.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man is sitting inside a restaurant while engaged in conversation with another individual, the scene is observed through the window with people walking by outside. The moment is captured with a sense of intimacy.
4.796463
1,920
810
115
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Mr__Rogers_Comes_to_Visit_Scene_(9_10)_|_Movieclips
4.603122
{'frame_00:00.png': 4.515364646911621, 'frame_00:01.png': 4.523961544036865, 'frame_00:02.png': 4.801211833953857, 'frame_00:03.png': 4.66363000869751, 'frame_00:04.png': 4.511440753936768}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Mr__Rogers_Comes_to_Visit_Scene_(9_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Mr__Rogers_Comes_to_Visit_Scene_(9_10)_|_Movieclips/short_scene_001.mp4
0.322792
1.060905
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Balanced'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Saturated, Yellow'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Left heavy'}]}
7.8
3.8
0
{"0": {"raw_counts": {"person": 6, "bed": 1, "couch": 1, "clock": 1, "chair": 1, "total": 10}, "bucketed_counts": {"humans": 6, "animals": 0, "vehicles": 0, "food": 0, "others": 4, "total": 10}}, "1": {"raw_counts": {"person": 4, "chair": 2, "couch": 1, "bed": 1, "cup": 1, "total": 9}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 5, "total": 9}}, "2": {"raw_counts": {"person": 3, "potted plant": 1, "chair": 2, "couch": 1, "bowl": 2, "total": 9}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 6, "total": 9}}, "3": {"raw_counts": {"vase": 1, "potted plant": 1, "person": 4, "bowl": 1, "total": 7}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 3, "total": 7}}, "4": {"raw_counts": {"vase": 1, "potted plant": 1, "person": 2, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}}
0.036321
[-1 1 1 1 1 1 1 -1 1 1 1 -1 1 -1 1 -1 1 -1 1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=b-VzKjKNPGM
scene_001
1
137
257
00:00:05.714
00:00:10.719
5.714047
10.719052
5.005005
5.714047
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A somber scene unfolds within a warmly lit room, where an elderly man reclines in a bed placed against a wall adorned with framed pictures and decorative vases. In the foreground, a person sits in a chair, their back to the camera, as they face the bed. To the right, a man sits in a chair by a table holding an old stereo system, gazing intently. A woman stands behind him, holding a glass, her expression a mix of concern and sadness. As the scene progresses, the woman moves slightly, her gaze fixed on the scene unfolding before her, while the seated man remains still, observing with a heavy heart. The room is filled with personal touches, from the bookshelf stocked with various items to the floral arrangement on the table, creating a lived-in atmosphere amidst the quiet drama.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A family gathers around a sick or elderly person in a bed, some taking photos, while others stand by, looking concerned. The scene is set in a modestly decorated living room, creating a somber and intimate atmosphere.
5.005005
1,920
810
120
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Pitching_a_Tent_Scene_(2_10)_|_Movieclips
4.914569
{'frame_00:00.png': 4.945087432861328, 'frame_00:01.png': 4.8422932624816895, 'frame_00:02.png': 5.008208274841309, 'frame_00:03.png': 4.870199203491211, 'frame_00:04.png': 4.907055377960205}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Pitching_a_Tent_Scene_(2_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Pitching_a_Tent_Scene_(2_10)_|_Movieclips/short_scene_001.mp4
-0.28894
0.233083
{'frame_00:00.png': [{'<COLOR>': 'Green, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Green, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Green, Blue'}, {'<LIGHTING>': 'Hard light, Top light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Green, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Green, Blue'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Right heavy'}]}
3.6
2
0
{"0": {"raw_counts": {"person": 2, "umbrella": 1, "tie": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "1": {"raw_counts": {"person": 2, "tie": 1, "umbrella": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "2": {"raw_counts": {"person": 2, "umbrella": 1, "tie": 1, "total": 4}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 2, "total": 4}}, "3": {"raw_counts": {"person": 2, "tie": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}, "4": {"raw_counts": {"person": 2, "tie": 1, "total": 3}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 3}}}
-0.02936
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 -1 1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=4OWBufqkWWw
scene_001
1
286
387
00:00:11.929
00:00:16.141
11.928595
16.141141
4.212546
11.928595
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A person in a maroon cardigan and tie stands next to a small, gray tent with red stripes, smiling and gesturing toward it, on what appears to be a television set with a blue backdrop showing green rolling hills; a camera operator stands to the left, partially visible, operating a large camera; ivy and greenery decorate the right side of the set, adding a touch of natural ambiance to the otherwise artificial environment. The man smiles, then walks towards the camera while taking off his cardigan.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man smiles as he stands next to a tent and looks at the camera while a camera operator films him. The scene is set on a set with a backdrop of hills and greenery.
4.212546
1,920
810
101
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Special_Puppet_Friends_Scene_(5_10)_|_Movieclips
4.216729
{'frame_00:00.png': 4.167184352874756, 'frame_00:01.png': 4.10935115814209, 'frame_00:02.png': 3.8560256958007812, 'frame_00:03.png': 4.411328315734863, 'frame_00:04.png': 4.539755344390869}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Special_Puppet_Friends_Scene_(5_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Special_Puppet_Friends_Scene_(5_10)_|_Movieclips/short_scene_001.mp4
-0.056091
0.116997
{'frame_00:00.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Warm, Desaturated'}, {'<LIGHTING>': 'Soft light, Top light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}]}
0.8
0.6
0.2
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "4": {"raw_counts": {"bird": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 1, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.083939
[-1 -1 -1 1 -1 -1 1 -1 -1 -1 -1 -1 -1 -1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1]
https://youtube.com/watch?v=XL6HMWkljm8
scene_001
1
175
276
00:00:07.299
00:00:11.512
7.298966
11.511512
4.212546
7.298966
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A hand opens a dark suitcase lined with a matching dark fabric. The fabric is lifted to reveal a stuffed animal cat with bright green eyes and a red nose lying next to a toy figure of a king with a white beard, crown, and royal robes. The hand carefully adjusts the toys within the suitcase.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A hand opens a suitcase to reveal a stuffed lion and a puppet resembling a king. These items are packed inside the black suitcase, suggesting someone is preparing for a trip.
4.212546
1,920
810
101
1920x810
short_scene_001_short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Would_You_Be_Mine?_Scene_(1_10)_|_Movieclips
4.15945
{'frame_00:00.png': 3.9910888671875, 'frame_00:01.png': 4.155650615692139, 'frame_00:02.png': 3.91422700881958, 'frame_00:03.png': 4.475545406341553, 'frame_00:04.png': 4.260739326477051}
short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Would_You_Be_Mine?_Scene_(1_10)_|_Movieclips
../assets/split_scenes/short_A_Beautiful_Day_in_the_Neighborhood_(2019)_-_Would_You_Be_Mine?_Scene_(1_10)_|_Movieclips/short_scene_001.mp4
-0.32026
0.619134
{'frame_00:00.png': [{'<COLOR>': 'Mixed, Saturated, Orange'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Daylight, Sunny'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:01.png': [{'<COLOR>': 'Mixed, Saturated, Orange, Yellow, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:02.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Cyan, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Center, Symmetrical'}], 'frame_00:03.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}], 'frame_00:04.png': [{'<COLOR>': 'Mixed, Saturated, Yellow, Blue'}, {'<LIGHTING>': 'Soft light'}, {'<LIGHTING_TYPE>': 'Artificial light, Practical light'}, {'<COMPOSITION>': 'Center'}]}
0.4
0
0
{"0": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "1": {"raw_counts": {"tennis racket": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}, "2": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "3": {"raw_counts": {"total": 0}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 0}}, "4": {"raw_counts": {"traffic light": 1, "total": 1}, "bucketed_counts": {"humans": 0, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 1}}}
-0.027352
[-1 1 1 1 -1 1 1 -1 -1 1 -1 -1 1 -1 1 -1 1 -1 -1 -1 -1 1 1 -1 -1 1]
https://youtube.com/watch?v=WVxVM2Lo5NM
scene_001
1
442
562
00:00:18.435
00:00:23.440
18.435102
23.440107
5.005005
18.435102
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A model home is seen through a circular, patterned lens. The lens shifts to reveal a yellow traffic light with a similar circular, patterned lens. The color and lens pattern remain constant as the view alternates between the model home and the traffic light. The traffic light is encased in a yellow frame and has bolts securing the lens. The background of the traffic light shifts slightly, showing a hint of blue and a framed picture. The scene concludes with the traffic light filling the frame, its yellow color and lens pattern consistent throughout the sequence.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A variety of shots show a yellow traffic light illuminated against a background. The light is filmed up close and shown in different lighting and angles.
5.005005
1,920
810
120
1920x810
short_scene_001_short_A_Bit_Nipply_Out_-_Christmas_Vacation_(4_10)_Movie_CLIP_(1989)_HD
4.285006
{'frame_00:00.png': 4.234274387359619, 'frame_00:01.png': 4.127614498138428, 'frame_00:02.png': 4.195295810699463, 'frame_00:03.png': 4.582840442657471}
short_A_Bit_Nipply_Out_-_Christmas_Vacation_(4_10)_Movie_CLIP_(1989)_HD
../assets/split_scenes/short_A_Bit_Nipply_Out_-_Christmas_Vacation_(4_10)_Movie_CLIP_(1989)_HD/short_scene_001.mp4
0.000723
1.031346
{'frame_00:00.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Left heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Center'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated'}, {'<LIGHTING>': 'Soft light, Low contrast, Side light'}, {'<LIGHTING_TYPE>': 'Daylight, Artificial light, Practical light'}, {'<COMPOSITION>': 'Right heavy, Short side'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Cyan'}, {'<LIGHTING>': 'Soft light, Side light'}, {'<LIGHTING_TYPE>': 'Daylight'}, {'<COMPOSITION>': 'Left heavy'}]}
1.5
1.5
0
{"0": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "1": {"raw_counts": {"person": 2, "total": 2}, "bucketed_counts": {"humans": 2, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 2}}, "2": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}, "3": {"raw_counts": {"person": 1, "total": 1}, "bucketed_counts": {"humans": 1, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 1}}}
-0.02983
[-1, 1, 1, -1, -1, 1, 1, -1, -1, 1, -1, -1, 1, -1, 1, -1, 1, -1, -1, -1, -1, 1, 1, -1, -1, 1]
https://youtube.com/watch?v=GxGkcC1VrhU
scene_001
1
59
149
00:00:02.461
00:00:06.215
2.460794
6.214548
3.753754
2.460794
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene focuses on a man in a light tan coat as he leans forward. His face is angled downward as he seems to examine something closely. As the scene progresses, the camera angle shifts slightly upward, revealing the man's face more clearly. He then looks up and speaks, a small smile appearing on his face. The background of the scene includes soft, out-of-focus lights, adding a gentle glow to the atmosphere.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
A man wearing a tan coat bends over to examine an object covered by a cloth; the man then looks up and speaks with an expression of interest. Soft Christmas lights are visible in the background.
3.753754
1,280
540
90
1280x540
short_scene_001_short_A_Boy's_Best_Friend_-_Psycho_(2_12)_Movie_CLIP_(1960)_HD
4.994056
{'frame_00:00.png': 5.0411577224731445, 'frame_00:01.png': 4.976540565490723, 'frame_00:02.png': 5.024566650390625, 'frame_00:03.png': 5.038547515869141, 'frame_00:04.png': 4.889469623565674}
short_A_Boy's_Best_Friend_-_Psycho_(2_12)_Movie_CLIP_(1960)_HD
../assets/split_scenes/short_A_Boy's_Best_Friend_-_Psycho_(2_12)_Movie_CLIP_(1960)_HD/short_scene_001.mp4
0.03857
0.115755
{'frame_00:00.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:01.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:02.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:03.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}], 'frame_00:04.png': [{'<COLOR>': 'Desaturated, Black and White'}, {'<LIGHTING>': 'Hard light, Side light'}, {'<LIGHTING_TYPE>': 'Artificial light'}, {'<COMPOSITION>': 'Right heavy'}]}
3.8
3.2
0
{"0": {"raw_counts": {"person": 3, "vase": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "1": {"raw_counts": {"person": 3, "vase": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "2": {"raw_counts": {"person": 3, "vase": 1, "total": 4}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 1, "total": 4}}, "3": {"raw_counts": {"person": 3, "total": 3}, "bucketed_counts": {"humans": 3, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 3}}, "4": {"raw_counts": {"person": 4, "total": 4}, "bucketed_counts": {"humans": 4, "animals": 0, "vehicles": 0, "food": 0, "others": 0, "total": 4}}}
0.006052
[-1, 1, 1, -1, -1, 1, -1, -1, -1, 1, -1, 1, 1, -1, 1, -1, -1, -1, 1, -1, -1, 1, 1, -1, -1, 1]
https://youtube.com/watch?v=I9mJ2oBONug
scene_001
1
43
163
00:00:01.793
00:00:06.798
1.79346
6.798465
5.005005
1.79346
You're a helpful assistant who is a part of a team working on video generation. As an assistant your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. Keeping that in mind, provide a caption for the input video. Please focus on the articulate details of the scenes presented in the video. Here are some guidelines: * Describe the composition of the scene, how it's progressing with all the components involved. * First describe the main subjects of the video and then how they are connected with one another. * DO NOT start the caption with "In this video,". * When appropriate for the scene, incorporate cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" IMPORTANT: Use these example phrases only when they naturally fit the scene you're describing. ## REQUIRED OUTPUT FORMAT: - Focus exclusively on what is visually happening in the frames - Describe actions, emotions, settings, and atmosphere - Maintain an objective, descriptive tone - Narrate as if describing the scene to someone who cannot see it - DO NOT mention character names, actor names, or film titles - Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The scene is set indoors, featuring a young man who is sitting and gesturing with his right arm raised, pointing towards the upper left of the frame. He is wearing a collared shirt and a jacket. Behind him, a wooden dresser is visible, topped with a decorative arrangement of what appear to be taxidermied animals, including a bird, positioned next to a pair of ornate candlesticks. The image is presented in black and white, with the lighting emphasizing the textures of the objects and the man's facial expressions. The overall composition suggests a staged setting, possibly a scene from a film, with attention to detail in the set design and the subject's pose.
You're a helpful assistant who is a part of a team working on video generation. As an assistant, your task is to provide a meaningful caption for a provided video. These captions will be used to fine-tune other video generation models. **IMPORTANT: Your caption must be no more than one to two sentences.** Keep it short, high-level, and descriptive—focusing on the overall action, mood, or atmosphere in the scene. Here are some guidelines: * Identify the main subjects and describe the general visual activity or setting. * Avoid going into fine-grained or overly specific details—this is a summary, not a full narration. * Use cinematic phrases such as: * "with dramatic lighting" * "in slow motion" * "captured in close-up" only when they naturally fit the visual tone of the video. ## REQUIRED OUTPUT FORMAT: * Focus exclusively on what is visually happening in the frames * Describe actions, emotions, settings, and atmosphere * Maintain an objective, descriptive tone * **Limit your response to a maximum of two sentences** * DO NOT mention character names, actor names, or film titles * DO NOT start the caption with "In this video," * Write in a continuous, cohesive paragraph that reads like a natural description of the entire scene, rather than listing or breaking it down frame by frame
The man sits in a dimly lit room with taxidermied animals, gesturing to the side while speaking. The black and white image evokes a sense of mystery.
5.005005
1,280
540
120
1280x540