LiteHRNet: Optimized for Qualcomm Devices

LiteHRNet is a machine learning model that detects human pose and returns a location and confidence for each of 17 joints.

This is based on the implementation of LiteHRNet 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 2.50, ONNX Runtime 1.30.0	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	Download

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

Model Details

Model Type: Model_use_case.pose_estimation

Model Stats:

• Input resolution: 256x192
• Model size (float): 4.49 MB
• Number of parameters: 1.11M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
LiteHRNet	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.315 ms	1 - 102 MB	NPU
LiteHRNet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	2.559 ms	0 - 100 MB	NPU
LiteHRNet	ONNX	float	Snapdragon® X2 Elite	2.412 ms	2 - 2 MB	NPU
LiteHRNet	ONNX	float	Snapdragon® X Elite	5.085 ms	5 - 5 MB	NPU
LiteHRNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	2.986 ms	0 - 125 MB	NPU
LiteHRNet	ONNX	float	Snapdragon® 8 Gen 1 Mobile	5.948 ms	1 - 122 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	4.347 ms	1 - 5 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.81 ms	0 - 8 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® QCS8450	5.948 ms	1 - 122 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	5.2 ms	1 - 4 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	5.085 ms	5 - 5 MB	NPU
LiteHRNet	ONNX	float	Qualcomm® Dragonwing™ Q-8750	2.559 ms	0 - 100 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.908 ms	1 - 85 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.074 ms	0 - 78 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® X2 Elite	1.31 ms	1 - 1 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® X Elite	2.41 ms	1 - 1 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.395 ms	0 - 100 MB	NPU
LiteHRNet	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	2.911 ms	0 - 100 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	2.181 ms	1 - 4 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.091 ms	1 - 2 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® SA8775P	2.68 ms	1 - 79 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® SA8650P	2.68 ms	1 - 79 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® SA8255P	2.68 ms	1 - 79 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® QCS8450	2.911 ms	0 - 100 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	2.476 ms	1 - 3 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	2.41 ms	1 - 1 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	1.074 ms	0 - 78 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® SA7255P	4.981 ms	1 - 78 MB	NPU
LiteHRNet	QNN_DLC	float	Qualcomm® SA8295P	3.348 ms	0 - 76 MB	NPU
LiteHRNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.049 ms	0 - 109 MB	NPU
LiteHRNet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	2.225 ms	0 - 106 MB	NPU
LiteHRNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.673 ms	0 - 146 MB	NPU
LiteHRNet	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	5.224 ms	0 - 135 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	4.198 ms	1 - 12 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.255 ms	1 - 5 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® SA8775P	5.247 ms	1 - 112 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® SA8650P	5.247 ms	1 - 112 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® SA8255P	5.247 ms	1 - 112 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® QCS8450	5.224 ms	0 - 135 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	4.91 ms	1 - 12 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	2.225 ms	0 - 106 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® SA7255P	8.564 ms	1 - 110 MB	NPU
LiteHRNet	TFLITE	float	Qualcomm® SA8295P	6.011 ms	1 - 111 MB	NPU

License

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

References

• Lite-HRNet: A Lightweight High-Resolution Network
• 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.