Hugging Face's logo

qualcomm
/
HRNetPose

Keypoint Detection
PyTorch
bu_auto
android
Model card Files
xet
 Community

HRNetPose: Optimized for Qualcomm Devices

HRNet performs pose estimation in high-resolution representations.

This is based on the implementation of HRNetPose 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 1.24.3 Download
ONNX w8a16 Universal QAIRT 2.42, ONNX Runtime 1.24.3 Download
ONNX w8a8 Universal QAIRT 2.42, ONNX Runtime 1.24.3 Download
QNN_DLC float Universal QAIRT 2.43 Download
QNN_DLC w8a16 Universal QAIRT 2.43 Download
QNN_DLC w8a8 Universal QAIRT 2.43 Download
TFLITE float Universal QAIRT 2.43, TFLite 2.19.1 Download
TFLITE w8a8 Universal QAIRT 2.43, TFLite 2.19.1 Download

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

Model Details

Model Type: Model_use_case.pose_estimation

Model Stats:

  • Model checkpoint: hrnet_posenet_FP32_state_dict
  • Input resolution: 256x192
  • Number of parameters: 28.5M
  • Model size (float): 109 MB
  • Model size (w8a8): 28.1 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
HRNetPose ONNX float Snapdragon® 8 Elite Gen 5 Mobile 1.218 ms 1 - 92 MB NPU
HRNetPose ONNX float Snapdragon® X2 Elite 1.335 ms 55 - 55 MB NPU
HRNetPose ONNX float Snapdragon® X Elite 2.618 ms 55 - 55 MB NPU
HRNetPose ONNX float Snapdragon® 8 Gen 3 Mobile 1.861 ms 0 - 147 MB NPU
HRNetPose ONNX float Qualcomm® QCS8550 (Proxy) 2.462 ms 0 - 58 MB NPU
HRNetPose ONNX float Qualcomm® QCS9075 3.913 ms 0 - 4 MB NPU
HRNetPose ONNX float Snapdragon® 8 Elite For Galaxy Mobile 1.5 ms 0 - 90 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.789 ms 0 - 128 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X2 Elite 1.008 ms 28 - 28 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X Elite 2.009 ms 28 - 28 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 1.255 ms 0 - 202 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS6490 487.565 ms 29 - 33 MB CPU
HRNetPose ONNX w8a16 Qualcomm® QCS8550 (Proxy) 1.722 ms 0 - 37 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS9075 2.019 ms 0 - 3 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCM6690 220.384 ms 31 - 49 MB CPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.971 ms 0 - 122 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 212.873 ms 26 - 41 MB CPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.804 ms 0 - 123 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X2 Elite 0.772 ms 30 - 30 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X Elite 1.634 ms 28 - 28 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 1.049 ms 0 - 187 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS6490 90.444 ms 8 - 68 MB CPU
HRNetPose ONNX w8a8 Qualcomm® QCS8550 (Proxy) 1.47 ms 0 - 5 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS9075 1.53 ms 0 - 3 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCM6690 61.642 ms 9 - 33 MB CPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.909 ms 0 - 118 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 59.026 ms 8 - 28 MB CPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.244 ms 1 - 78 MB NPU
HRNetPose QNN_DLC float Snapdragon® X2 Elite 1.659 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® X Elite 2.944 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.976 ms 0 - 118 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8275 (Proxy) 14.164 ms 1 - 75 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8550 (Proxy) 2.726 ms 1 - 2 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA8775P 4.307 ms 1 - 76 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS9075 4.134 ms 3 - 5 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8450 (Proxy) 4.953 ms 0 - 107 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA7255P 14.164 ms 1 - 75 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA8295P 4.592 ms 1 - 68 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 1.55 ms 0 - 73 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.805 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X2 Elite 1.277 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X Elite 2.151 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.4 ms 0 - 152 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS6490 6.756 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8275 (Proxy) 5.171 ms 0 - 98 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8550 (Proxy) 1.91 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA8775P 2.294 ms 0 - 101 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS9075 2.208 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCM6690 21.69 ms 0 - 217 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8450 (Proxy) 2.712 ms 0 - 151 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA7255P 5.171 ms 0 - 98 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA8295P 3.151 ms 0 - 97 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 1.012 ms 0 - 101 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 2.544 ms 0 - 99 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.571 ms 0 - 93 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X2 Elite 0.738 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X Elite 1.264 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.835 ms 0 - 133 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS6490 3.736 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8275 (Proxy) 2.83 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 1.145 ms 0 - 43 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA8775P 1.519 ms 0 - 91 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS9075 1.326 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCM6690 10.249 ms 0 - 212 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8450 (Proxy) 1.771 ms 0 - 136 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA7255P 2.83 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA8295P 2.004 ms 0 - 87 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.645 ms 0 - 92 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 1.519 ms 0 - 88 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 1.286 ms 0 - 118 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Gen 3 Mobile 1.976 ms 0 - 186 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8275 (Proxy) 14.212 ms 0 - 114 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8550 (Proxy) 2.772 ms 0 - 3 MB NPU
HRNetPose TFLITE float Qualcomm® SA8775P 4.34 ms 0 - 117 MB NPU
HRNetPose TFLITE float Qualcomm® QCS9075 4.152 ms 0 - 58 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8450 (Proxy) 5.027 ms 0 - 178 MB NPU
HRNetPose TFLITE float Qualcomm® SA7255P 14.212 ms 0 - 114 MB NPU
HRNetPose TFLITE float Qualcomm® SA8295P 4.625 ms 0 - 106 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 1.571 ms 0 - 118 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.515 ms 0 - 90 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.724 ms 0 - 143 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS6490 3.339 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8275 (Proxy) 2.597 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 0.951 ms 0 - 20 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8775P 1.376 ms 0 - 89 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS9075 1.113 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCM6690 9.595 ms 0 - 207 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8450 (Proxy) 1.551 ms 0 - 142 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA7255P 2.597 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8295P 1.812 ms 0 - 85 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.581 ms 0 - 90 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 1.354 ms 0 - 85 MB NPU

License

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

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Keypoint Detection
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/HRNetPose