You can deploy this model and start earning money today!

qualcomm/
EyeGaze

Views

No views yet

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	QAIRT 2.45, ONNX Runtime 1.27.1	Download
ONNX	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	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:

Input resolution: 96x160
Model checkpoint: checkpoint.pt
Model size (float): 9.6MB
Model size (w8a16): 3.3 MB
Number of parameters: 2.58M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EyeGaze	ONNX	float	Snapdragon® X2 Elite	0.613 ms	8 - 8 MB	NPU
EyeGaze	ONNX	float	Snapdragon® X Elite	1.285 ms	8 - 8 MB	NPU
EyeGaze	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.871 ms	0 - 64 MB	NPU
EyeGaze	ONNX	float	Snapdragon® 8 Gen 1 Mobile	1.725 ms	2 - 67 MB	NPU
EyeGaze	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	1.704 ms	0 - 4 MB	NPU
EyeGaze	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.247 ms	0 - 30 MB	NPU
EyeGaze	ONNX	float	Qualcomm® QCS8450	1.725 ms	2 - 67 MB	NPU
EyeGaze	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	1.711 ms	0 - 3 MB	NPU
EyeGaze	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	1.285 ms	8 - 8 MB	NPU
EyeGaze	ONNX	float	Qualcomm® Dragonwing™ Q-8750	0.688 ms	0 - 51 MB	NPU
EyeGaze	ONNX	float	Snapdragon® 8 Elite Mobile	0.688 ms	0 - 51 MB	NPU
EyeGaze	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.597 ms	0 - 52 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® X2 Elite	0.526 ms	3 - 3 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® X Elite	0.977 ms	0 - 0 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.635 ms	0 - 80 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	1.14 ms	0 - 79 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	3.877 ms	2 - 4 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	0.985 ms	0 - 4 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.963 ms	0 - 6 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® QCS8450	1.14 ms	0 - 79 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	1.115 ms	0 - 3 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	0.977 ms	0 - 0 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	6.154 ms	2 - 184 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	1.132 ms	0 - 180 MB	NPU
EyeGaze	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	0.514 ms	0 - 69 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® 8 Elite Mobile	0.514 ms	0 - 69 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.463 ms	0 - 65 MB	NPU
EyeGaze	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	1.132 ms	0 - 180 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® X2 Elite	0.711 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® X Elite	1.168 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.738 ms	0 - 52 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	1.508 ms	0 - 55 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	1.475 ms	0 - 4 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	4.187 ms	0 - 39 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.015 ms	0 - 10 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® SA8775P	1.634 ms	0 - 41 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® SA8650P	1.634 ms	0 - 41 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® SA8255P	1.634 ms	0 - 41 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® QCS8450	1.508 ms	0 - 55 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	1.475 ms	0 - 3 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	1.168 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	0.595 ms	0 - 43 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® SA7255P	4.187 ms	0 - 39 MB	NPU
EyeGaze	QNN_DLC	float	Qualcomm® SA8295P	1.812 ms	0 - 38 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® 8 Elite Mobile	0.595 ms	0 - 43 MB	NPU
EyeGaze	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.545 ms	0 - 39 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.822 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® X Elite	1.491 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	0.886 ms	0 - 59 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® 8 Gen 1 Mobile	1.894 ms	0 - 61 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	1.263 ms	0 - 3 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	2.795 ms	0 - 46 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.313 ms	0 - 13 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® SA8775P	1.622 ms	0 - 48 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® SA8650P	1.622 ms	0 - 48 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® SA8255P	1.622 ms	0 - 48 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® QCS8450	1.894 ms	0 - 61 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	1.452 ms	2 - 4 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	1.491 ms	0 - 0 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	0.674 ms	0 - 50 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® SA7255P	2.795 ms	0 - 46 MB	NPU
EyeGaze	QNN_DLC	w8a16	Qualcomm® SA8295P	2.385 ms	0 - 44 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	0.674 ms	0 - 50 MB	NPU
EyeGaze	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.544 ms	0 - 44 MB	NPU
EyeGaze	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.977 ms	0 - 52 MB	NPU
EyeGaze	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	1.886 ms	0 - 58 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	1.821 ms	0 - 10 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	4.765 ms	0 - 40 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.402 ms	0 - 3 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® SA8775P	2.071 ms	0 - 43 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® SA8650P	2.071 ms	0 - 43 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® SA8255P	2.071 ms	0 - 43 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® QCS8450	1.886 ms	0 - 58 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	1.866 ms	0 - 9 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	0.785 ms	0 - 45 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® SA7255P	4.765 ms	0 - 40 MB	NPU
EyeGaze	TFLITE	float	Qualcomm® SA8295P	2.291 ms	0 - 38 MB	NPU
EyeGaze	TFLITE	float	Snapdragon® 8 Elite Mobile	0.785 ms	0 - 45 MB	NPU
EyeGaze	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.687 ms	0 - 39 MB	NPU

License

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

References

Source Model Implementation

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.

qualcomm/EyeGaze