CenterNet-Pose: Optimized for Qualcomm Devices

CenterNet-Pose 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 CenterNet-Pose 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
PRECOMPILED_QNN_ONNX	float	Snapdragon® X2 Elite	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® X Elite	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 3 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 1 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8550 (Proxy)	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS9075	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_CONTEXT_BINARY	float	Snapdragon® X2 Elite	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® X Elite	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 3 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 1 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® QCS8550 (Proxy)	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA8775P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Gen 5 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA7255P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA8295P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® QCS9075	QAIRT 2.45	Download

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

Model Details

Model Type: Model_use_case.pose_estimation

Model Stats:

• Model checkpoint: multi_pose_dla_3x.pth
• Input resolution: 1 x 3 x 512 x 512
• Number of parameters: 20.6M
• Model size: 57.8 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® X2 Elite	27.545 ms	4 - 4 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® X Elite	57.442 ms	44 - 44 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 3 Mobile	38.774 ms	4 - 10 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 1 Mobile	88.99 ms	4 - 17 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8550 (Proxy)	56.124 ms	0 - 49 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8450	88.99 ms	4 - 17 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	26.852 ms	4 - 11 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS9075	58.836 ms	3 - 6 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Mobile	30.284 ms	4 - 11 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8750	30.284 ms	4 - 11 MB	NPU
CenterNet-Pose	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS7181	57.442 ms	44 - 44 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® X2 Elite	27.703 ms	1 - 1 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® X Elite	57.598 ms	1 - 1 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 3 Mobile	38.831 ms	1 - 8 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 1 Mobile	91.222 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8275	104.19 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8550 (Proxy)	56.194 ms	2 - 3 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8450	91.222 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Gen 5 Mobile	26.853 ms	1 - 9 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® SA7255P	104.19 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS9075	58.347 ms	1 - 4 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Mobile	30.644 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® SA8295P	79.761 ms	0 - 6 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8750	30.644 ms	1 - 10 MB	NPU
CenterNet-Pose	QNN_CONTEXT_BINARY	float	Qualcomm® QCS7181	57.598 ms	1 - 1 MB	NPU

License

• The license for the original implementation of CenterNet-Pose can be found here.

References

• Objects as Points
• 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.