MediaPipe-Hand-Detection: Optimized for Qualcomm Devices

The MediaPipe Hand Landmark Detector is a machine learning pipeline that predicts bounding boxes and pose skeletons of hands in an image.

This is based on the implementation of MediaPipe-Hand-Detection 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 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download

For more device-specific assets and performance metrics, visit MediaPipe-Hand-Detection 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 MediaPipe-Hand-Detection on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 256x256
• Number of parameters (HandDetector): 1.76M
• Model size (HandDetector) (float): 6.75 MB
• Number of parameters (HandLandmarkDetector): 2.01M
• Model size (HandLandmarkDetector) (float): 7.70 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
HandDetector	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.42 ms	0 - 47 MB	NPU
HandDetector	ONNX	float	Snapdragon® X2 Elite	0.489 ms	0 - 0 MB	NPU
HandDetector	ONNX	float	Snapdragon® X Elite	1.019 ms	3 - 3 MB	NPU
HandDetector	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.595 ms	0 - 74 MB	NPU
HandDetector	ONNX	float	Qualcomm® QCS8550 (Proxy)	0.843 ms	0 - 7 MB	NPU
HandDetector	ONNX	float	Qualcomm® QCS9075	1.267 ms	1 - 3 MB	NPU
HandDetector	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.499 ms	0 - 49 MB	NPU
HandDetector	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.378 ms	1 - 44 MB	NPU
HandDetector	QNN_DLC	float	Snapdragon® X2 Elite	0.628 ms	1 - 1 MB	NPU
HandDetector	QNN_DLC	float	Snapdragon® X Elite	0.926 ms	1 - 1 MB	NPU
HandDetector	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.537 ms	0 - 60 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	3.819 ms	1 - 40 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	0.736 ms	1 - 2 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® SA8775P	5.169 ms	1 - 40 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® QCS9075	1.13 ms	1 - 3 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	1.389 ms	0 - 54 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® SA7255P	3.819 ms	1 - 40 MB	NPU
HandDetector	QNN_DLC	float	Qualcomm® SA8295P	1.691 ms	0 - 31 MB	NPU
HandDetector	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.446 ms	0 - 40 MB	NPU
HandDetector	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.381 ms	0 - 42 MB	NPU
HandDetector	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.535 ms	0 - 58 MB	NPU
HandDetector	TFLITE	float	Qualcomm® QCS8275 (Proxy)	3.831 ms	0 - 38 MB	NPU
HandDetector	TFLITE	float	Qualcomm® QCS8550 (Proxy)	0.738 ms	0 - 2 MB	NPU
HandDetector	TFLITE	float	Qualcomm® SA8775P	1.323 ms	0 - 42 MB	NPU
HandDetector	TFLITE	float	Qualcomm® QCS9075	1.144 ms	0 - 7 MB	NPU
HandDetector	TFLITE	float	Qualcomm® QCS8450 (Proxy)	1.399 ms	0 - 51 MB	NPU
HandDetector	TFLITE	float	Qualcomm® SA7255P	3.831 ms	0 - 38 MB	NPU
HandDetector	TFLITE	float	Qualcomm® SA8295P	1.699 ms	0 - 30 MB	NPU
HandDetector	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.454 ms	0 - 44 MB	NPU
HandLandmarkDetector	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.563 ms	0 - 43 MB	NPU
HandLandmarkDetector	ONNX	float	Snapdragon® X2 Elite	0.716 ms	6 - 6 MB	NPU
HandLandmarkDetector	ONNX	float	Snapdragon® X Elite	1.413 ms	6 - 6 MB	NPU
HandLandmarkDetector	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.846 ms	0 - 69 MB	NPU
HandLandmarkDetector	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.113 ms	0 - 9 MB	NPU
HandLandmarkDetector	ONNX	float	Qualcomm® QCS9075	1.843 ms	1 - 4 MB	NPU
HandLandmarkDetector	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.659 ms	0 - 44 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.516 ms	1 - 39 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Snapdragon® X2 Elite	0.856 ms	1 - 1 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Snapdragon® X Elite	1.284 ms	1 - 1 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.769 ms	0 - 61 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	5.344 ms	1 - 35 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.007 ms	1 - 2 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® SA8775P	1.871 ms	1 - 39 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® QCS9075	1.696 ms	1 - 3 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	1.934 ms	0 - 52 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® SA7255P	5.344 ms	1 - 35 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Qualcomm® SA8295P	2.245 ms	0 - 31 MB	NPU
HandLandmarkDetector	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.601 ms	0 - 34 MB	NPU
HandLandmarkDetector	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.513 ms	0 - 44 MB	NPU
HandLandmarkDetector	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.776 ms	0 - 60 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® QCS8275 (Proxy)	5.345 ms	0 - 40 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.0 ms	0 - 79 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® SA8775P	1.869 ms	0 - 43 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® QCS9075	1.686 ms	0 - 9 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® QCS8450 (Proxy)	1.948 ms	0 - 57 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® SA7255P	5.345 ms	0 - 40 MB	NPU
HandLandmarkDetector	TFLITE	float	Qualcomm® SA8295P	2.242 ms	0 - 35 MB	NPU
HandLandmarkDetector	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.609 ms	0 - 45 MB	NPU

License

• The license for the original implementation of MediaPipe-Hand-Detection can be found here.

References

• MediaPipe Hands: On-device Real-time Hand Tracking
• 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.