RTMPose-Body2d: Optimized for Qualcomm Devices

RTMPose is a machine learning model that detects human pose and returns a location and confidence for each of 133 joints.

This is based on the implementation of RTMPose-Body2d 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
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a16	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 RTMPose-Body2d 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 RTMPose-Body2d on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.pose_estimation

Model Stats:

• Input resolution: 256x192
• Number of parameters: 17.9M
• Model size (float): 68.5 MB
• Model size (w8a16): 18.2 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
RTMPose-Body2d	ONNX	float	Snapdragon® X2 Elite	0.923 ms	37 - 37 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® X Elite	1.889 ms	37 - 37 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.334 ms	0 - 62 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.767 ms	0 - 42 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS9075	2.457 ms	1 - 4 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.094 ms	0 - 36 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.924 ms	0 - 39 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® X2 Elite	0.769 ms	19 - 19 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® X Elite	1.934 ms	19 - 19 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	1.168 ms	0 - 91 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS6490	178.11 ms	49 - 53 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	1.705 ms	0 - 22 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS9075	1.91 ms	0 - 3 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCM6690	86.948 ms	37 - 46 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.861 ms	0 - 59 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	82.847 ms	48 - 57 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.737 ms	0 - 60 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® X2 Elite	1.103 ms	1 - 1 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® X Elite	1.857 ms	1 - 1 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.328 ms	0 - 58 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	7.532 ms	1 - 33 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.737 ms	1 - 2 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8775P	2.432 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS9075	2.412 ms	1 - 3 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.539 ms	0 - 63 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA7255P	7.532 ms	1 - 33 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8295P	3.526 ms	0 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.075 ms	0 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.907 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.998 ms	0 - 0 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® X Elite	1.941 ms	0 - 0 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.185 ms	0 - 80 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	3.882 ms	0 - 49 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	1.734 ms	0 - 2 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA8775P	2.052 ms	0 - 52 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS9075	1.899 ms	0 - 2 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA7255P	3.882 ms	0 - 49 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.892 ms	0 - 50 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.757 ms	0 - 51 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.326 ms	0 - 90 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8275 (Proxy)	7.492 ms	0 - 49 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.755 ms	0 - 2 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8775P	2.419 ms	0 - 51 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS9075	2.415 ms	0 - 40 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.52 ms	0 - 97 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA7255P	7.492 ms	0 - 49 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8295P	3.563 ms	0 - 51 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.076 ms	0 - 48 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.9 ms	0 - 50 MB	NPU

License

• The license for the original implementation of RTMPose-Body2d can be found here.

References

• RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose
• 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.