update application file

app.py

@@ -1,7 +1,6 @@
 import gradio as gr
 import cv2
 import os
-import spaces
 import tempfile
 from torchvision import transforms
 from torchvision.transforms import Compose
@@ -9,12 +8,10 @@ import torch
 import numpy as np
 from PIL import Image
 import torch.nn.functional as F
-from pytorchvideo.transforms.functional import predict_depth
 from transformers import pipeline, TimesformerModel, VideoMAEImageProcessor
 from utils import *
 from algorithm import *
 
-@spaces.GPU
 def make_video(video_path, outdir='./summarized_video',encoder='Kmeans'):
     if encoder not in ["Kmeans", "Sum of Squared Difference 01", "Sum of Squared Difference 02"]:
         encoder = "Kmeans"

requirements.txt

@@ -1,7 +1,9 @@
 gradio_imageslider
-gradio==4.14.0
+gradio==4.29.0
 torch
 pillow
 torchvision
 opencv-python
-huggingface_hub
+huggingface_hub
+pytorchvideo==0.1.5
+spaces

utils.py

@@ -52,15 +52,6 @@ def to_video(selected_frames, frames, output_path, video_fps):
     video_writer.release()
     print("Completed summarizing the video (wait for a moment to load).")
 
-def to_txt(selected_frames, output_path, clip_sample_rate):
-    # Write the selected frames to a txt file
-
-    with open(output_path, "w") as file:
-        for item in selected_frames:
-            file.write(str(item) + "\n")
-    
-    print("Completed summarizing the txt (wait for a moment to load).")
-
 def load_model():
     try:
         DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'