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 1.25.0	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.25.0	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 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® 8 Elite Gen 5 Mobile	0.92 ms	0 - 38 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® X Elite	1.723 ms	149 - 149 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.341 ms	0 - 61 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.775 ms	0 - 52 MB	NPU
RTMPose-Body2d	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.106 ms	0 - 40 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS9075	2.437 ms	1 - 46 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS8750	1.106 ms	0 - 40 MB	NPU
RTMPose-Body2d	ONNX	float	Qualcomm® QCS7181	1.723 ms	149 - 149 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.739 ms	0 - 62 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® X2 Elite	0.789 ms	212 - 212 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® X Elite	1.761 ms	181 - 181 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	1.17 ms	0 - 93 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS6490	174.374 ms	47 - 49 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	1.714 ms	0 - 47 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCM6690	87.013 ms	38 - 46 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	82.728 ms	48 - 56 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.869 ms	0 - 64 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS9075	1.908 ms	0 - 45 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS7790	82.728 ms	48 - 56 MB	CPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS8750	0.869 ms	0 - 64 MB	NPU
RTMPose-Body2d	ONNX	w8a16	Qualcomm® QCS7181	1.761 ms	181 - 181 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.889 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® X2 Elite	1.054 ms	1 - 1 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® X Elite	1.85 ms	1 - 1 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.325 ms	0 - 55 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8275	7.473 ms	1 - 32 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.73 ms	1 - 2 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8775P	2.44 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8650P	2.44 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8255P	2.44 ms	1 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.503 ms	0 - 61 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA7255P	7.473 ms	1 - 32 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® SA8295P	3.538 ms	0 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.069 ms	0 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS9075	2.361 ms	3 - 5 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS8750	1.069 ms	0 - 35 MB	NPU
RTMPose-Body2d	QNN_DLC	float	Qualcomm® QCS7181	1.85 ms	1 - 1 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.741 ms	0 - 49 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.922 ms	0 - 0 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® X Elite	1.897 ms	0 - 0 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.199 ms	0 - 76 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS8275	3.86 ms	0 - 48 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	1.705 ms	0 - 11 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA8775P	2.049 ms	0 - 51 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA8650P	2.049 ms	0 - 51 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA8255P	2.049 ms	0 - 51 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® SA7255P	3.86 ms	0 - 48 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.88 ms	0 - 52 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS9075	1.922 ms	0 - 2 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS8750	0.88 ms	0 - 52 MB	NPU
RTMPose-Body2d	QNN_DLC	w8a16	Qualcomm® QCS7181	1.897 ms	0 - 0 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.886 ms	0 - 47 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.257 ms	0 - 76 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8275	7.488 ms	1 - 44 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.703 ms	0 - 3 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8775P	2.375 ms	0 - 45 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8650P	2.375 ms	0 - 45 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8255P	2.375 ms	0 - 45 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.512 ms	0 - 85 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA7255P	7.488 ms	1 - 44 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® SA8295P	3.501 ms	0 - 47 MB	NPU
RTMPose-Body2d	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.037 ms	0 - 47 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS9075	2.347 ms	0 - 40 MB	NPU
RTMPose-Body2d	TFLITE	float	Qualcomm® QCS8750	1.037 ms	0 - 47 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.