YAML Metadata Warning:The pipeline tag "gaze-estimation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
EyeGaze: Optimized for Qualcomm Devices
Predicts gaze direction (pitch, yaw) from 96x160 grayscale eye images using the EyeNet model.
This is based on the implementation of EyeGaze found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | ONNX Runtime 1.24.1 | Download |
| ONNX | w8a16 | Universal | ONNX Runtime 1.24.1 | Download |
| QNN_DLC | float | Universal | QAIRT 2.43 | Download |
| TFLITE | float | Universal | TFLite 2.17.0 | Download |
For more device-specific assets and performance metrics, visit EyeGaze on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for EyeGaze on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.gaze_estimation
Model Stats:
- Model checkpoint: checkpoint.pt
- Input resolution: 96x160
- Number of parameters: 2.58M
- Model size (float): 9.6MB
- Model size (w8a16): 3.3 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| EyeGaze | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 19.006 ms | 32 - 43 MB | CPU |
| EyeGaze | ONNX | float | Snapdragon® X2 Elite | 9.656 ms | 35 - 35 MB | CPU |
| EyeGaze | ONNX | float | Snapdragon® X Elite | 8.814 ms | 34 - 34 MB | CPU |
| EyeGaze | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 22.597 ms | 31 - 41 MB | CPU |
| EyeGaze | ONNX | float | Qualcomm® QCS8550 (Proxy) | 26.679 ms | 27 - 35 MB | CPU |
| EyeGaze | ONNX | float | Qualcomm® QCS9075 | 21.009 ms | 32 - 42 MB | CPU |
| EyeGaze | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 20.026 ms | 32 - 41 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 26.022 ms | 69 - 85 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® X2 Elite | 17.366 ms | 102 - 102 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® X Elite | 20.597 ms | 101 - 101 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 32.035 ms | 70 - 85 MB | CPU |
| EyeGaze | ONNX | w8a16 | Qualcomm® QCS6490 | 146.319 ms | 69 - 75 MB | CPU |
| EyeGaze | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 37.513 ms | 42 - 51 MB | CPU |
| EyeGaze | ONNX | w8a16 | Qualcomm® QCS9075 | 40.922 ms | 72 - 78 MB | CPU |
| EyeGaze | ONNX | w8a16 | Qualcomm® QCM6690 | 68.7 ms | 65 - 76 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 27.174 ms | 92 - 104 MB | CPU |
| EyeGaze | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 59.525 ms | 61 - 72 MB | CPU |
| EyeGaze | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 23.952 ms | 36 - 51 MB | CPU |
| EyeGaze | QNN_DLC | float | Snapdragon® X2 Elite | 36.429 ms | 14 - 14 MB | CPU |
| EyeGaze | QNN_DLC | float | Snapdragon® X Elite | 28.842 ms | 13 - 13 MB | CPU |
| EyeGaze | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 33.611 ms | 10 - 23 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 234.481 ms | 36 - 50 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 45.602 ms | 36 - 41 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® SA8775P | 50.604 ms | 34 - 43 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® QCS9075 | 56.349 ms | 61 - 133 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 65.932 ms | 11 - 25 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® SA7255P | 234.481 ms | 36 - 50 MB | CPU |
| EyeGaze | QNN_DLC | float | Qualcomm® SA8295P | 42.588 ms | 33 - 42 MB | CPU |
| EyeGaze | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 39.099 ms | 36 - 51 MB | CPU |
License
- The license for the original implementation of EyeGaze can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.