Hugging Face's logo

qualcomm
/
EyeGaze

PyTorch
bu_auto
real_time
android
gaze-estimation
Model card Files
xet
 Community

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 1.25.0 Download
ONNX w8a16 Universal ONNX Runtime 1.25.0 Download
QNN_DLC float Universal QAIRT 2.45 Download
TFLITE float Universal 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® X2 Elite 0.612 ms 8 - 8 MB NPU
EyeGaze ONNX float Snapdragon® X Elite 1.264 ms 8 - 8 MB NPU
EyeGaze ONNX float Snapdragon® 8 Gen 3 Mobile 0.872 ms 0 - 64 MB NPU
EyeGaze ONNX float Snapdragon® 8 Gen 1 Mobile 1.731 ms 0 - 65 MB NPU
EyeGaze ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.254 ms 0 - 9 MB NPU
EyeGaze ONNX float Qualcomm® QCS8450 1.731 ms 0 - 65 MB NPU
EyeGaze ONNX float Qualcomm® Dragonwing™ IQ-9075 1.71 ms 0 - 3 MB NPU
EyeGaze ONNX float Snapdragon® 8 Elite Gen 5 Mobile 0.607 ms 0 - 50 MB NPU
EyeGaze ONNX float Snapdragon® 8 Elite Mobile 0.69 ms 0 - 46 MB NPU
EyeGaze ONNX float Qualcomm® Dragonwing™ Q-8750 0.69 ms 0 - 46 MB NPU
EyeGaze ONNX float Qualcomm® Dragonwing™ IQ-X7181 1.264 ms 8 - 8 MB NPU
EyeGaze ONNX w8a16 Snapdragon® X2 Elite 0.561 ms 3 - 3 MB NPU
EyeGaze ONNX w8a16 Snapdragon® X Elite 1.013 ms 0 - 0 MB NPU
EyeGaze ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.659 ms 0 - 81 MB NPU
EyeGaze ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 45.46 ms 60 - 79 MB CPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.984 ms 0 - 6 MB NPU
EyeGaze ONNX w8a16 Qualcomm® QCS8450 45.46 ms 60 - 79 MB CPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 1.138 ms 0 - 3 MB NPU
EyeGaze ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 1.166 ms 0 - 66 MB NPU
EyeGaze ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.498 ms 0 - 68 MB NPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 6.66 ms 2 - 183 MB NPU
EyeGaze ONNX w8a16 Snapdragon® 8 Elite Mobile 0.55 ms 0 - 70 MB NPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 1.166 ms 0 - 66 MB NPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.55 ms 0 - 70 MB NPU
EyeGaze ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.013 ms 0 - 0 MB NPU
EyeGaze QNN_DLC float Snapdragon® X2 Elite 0.716 ms 0 - 0 MB NPU
EyeGaze QNN_DLC float Snapdragon® X Elite 1.186 ms 0 - 0 MB NPU
EyeGaze QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.734 ms 0 - 52 MB NPU
EyeGaze QNN_DLC float Snapdragon® 8 Gen 1 Mobile 1.504 ms 0 - 56 MB NPU
EyeGaze QNN_DLC float Qualcomm® QCS8275 4.176 ms 0 - 38 MB NPU
EyeGaze QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.011 ms 0 - 1 MB NPU
EyeGaze QNN_DLC float Qualcomm® SA8775P 1.621 ms 0 - 41 MB NPU
EyeGaze QNN_DLC float Qualcomm® SA8650P 1.621 ms 0 - 41 MB NPU
EyeGaze QNN_DLC float Qualcomm® SA8255P 1.621 ms 0 - 41 MB NPU
EyeGaze QNN_DLC float Qualcomm® QCS8450 1.504 ms 0 - 56 MB NPU
EyeGaze QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.469 ms 0 - 3 MB NPU
EyeGaze QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.549 ms 0 - 39 MB NPU
EyeGaze QNN_DLC float Qualcomm® SA7255P 4.176 ms 0 - 38 MB NPU
EyeGaze QNN_DLC float Snapdragon® 8 Elite Mobile 0.601 ms 0 - 42 MB NPU
EyeGaze QNN_DLC float Qualcomm® SA8295P 1.807 ms 0 - 38 MB NPU
EyeGaze QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.601 ms 0 - 42 MB NPU
EyeGaze QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.186 ms 0 - 0 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® X2 Elite 0.807 ms 0 - 0 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® X Elite 1.487 ms 0 - 0 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 0.884 ms 0 - 57 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 1.871 ms 0 - 61 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® QCS8275 2.779 ms 0 - 46 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.299 ms 0 - 4 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® SA8775P 1.621 ms 0 - 48 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® SA8650P 1.621 ms 0 - 48 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® SA8255P 1.621 ms 0 - 48 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® QCS8450 1.871 ms 0 - 61 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.446 ms 0 - 2 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.563 ms 0 - 48 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® SA7255P 2.779 ms 0 - 46 MB NPU
EyeGaze QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.678 ms 0 - 49 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® SA8295P 2.376 ms 0 - 44 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.678 ms 0 - 49 MB NPU
EyeGaze QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.487 ms 0 - 0 MB NPU
EyeGaze TFLITE float Snapdragon® 8 Gen 3 Mobile 1.264 ms 0 - 170 MB NPU
EyeGaze TFLITE float Snapdragon® 8 Gen 1 Mobile 2.922 ms 0 - 174 MB NPU
EyeGaze TFLITE float Qualcomm® QCS8275 5.287 ms 2 - 159 MB NPU
EyeGaze TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.801 ms 0 - 3 MB NPU
EyeGaze TFLITE float Qualcomm® SA8775P 2.492 ms 0 - 161 MB NPU
EyeGaze TFLITE float Qualcomm® SA8650P 2.492 ms 0 - 161 MB NPU
EyeGaze TFLITE float Qualcomm® SA8255P 2.492 ms 0 - 161 MB NPU
EyeGaze TFLITE float Qualcomm® QCS8450 2.922 ms 0 - 174 MB NPU
EyeGaze TFLITE float Qualcomm® Dragonwing™ IQ-9075 2.804 ms 0 - 11 MB NPU
EyeGaze TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.949 ms 0 - 160 MB NPU
EyeGaze TFLITE float Qualcomm® SA7255P 5.287 ms 2 - 159 MB NPU
EyeGaze TFLITE float Snapdragon® 8 Elite Mobile 1.102 ms 0 - 156 MB NPU
EyeGaze TFLITE float Qualcomm® SA8295P 2.875 ms 0 - 156 MB NPU
EyeGaze TFLITE float Qualcomm® Dragonwing™ Q-8750 1.102 ms 0 - 156 MB NPU

License

  • The license for the original implementation of EyeGaze can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support