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 2.45, ONNX Runtime 1.25.0 Download
ONNX w8a16 Universal QAIRT 2.45, ONNX Runtime 1.25.0 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.25.0 Download
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a16 Universal QAIRT 2.45 Download
QNN_DLC w8a8 Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download
TFLITE w8a8 Universal QAIRT 2.45 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® X2 Elite 1.321 ms 212 - 212 MB NPU
HRNetPose ONNX float Snapdragon® X Elite 2.428 ms 148 - 148 MB NPU
HRNetPose ONNX float Snapdragon® 8 Gen 3 Mobile 1.846 ms 1 - 131 MB NPU
HRNetPose ONNX float Snapdragon® 8 Gen 1 Mobile 4.322 ms 1 - 120 MB NPU
HRNetPose ONNX float Qualcomm® QCS8550 (Proxy) 2.484 ms 1 - 4 MB NPU
HRNetPose ONNX float Qualcomm® QCS8450 4.322 ms 1 - 120 MB NPU
HRNetPose ONNX float Snapdragon® 8 Elite Mobile 1.449 ms 0 - 86 MB NPU
HRNetPose ONNX float Snapdragon® 8 Elite Gen 5 Mobile 1.241 ms 0 - 84 MB NPU
HRNetPose ONNX float Qualcomm® QCS9075 3.895 ms 1 - 46 MB NPU
HRNetPose ONNX float Qualcomm® QCS8750 1.449 ms 0 - 86 MB NPU
HRNetPose ONNX float Qualcomm® QCS7181 2.428 ms 148 - 148 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X2 Elite 1.007 ms 213 - 213 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X Elite 1.753 ms 149 - 149 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 1.23 ms 0 - 189 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 2.458 ms 0 - 192 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS6490 6.206 ms 0 - 46 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS8550 (Proxy) 1.718 ms 0 - 4 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS8450 2.458 ms 0 - 192 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.799 ms 0 - 131 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite Mobile 0.973 ms 0 - 133 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS9075 1.984 ms 0 - 45 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 2.328 ms 0 - 132 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCM6690 18.986 ms 0 - 248 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS7790 2.328 ms 0 - 132 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS8750 0.973 ms 0 - 133 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS7181 1.753 ms 149 - 149 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X2 Elite 0.568 ms 213 - 213 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X Elite 1.117 ms 149 - 149 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 0.861 ms 0 - 178 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 1.737 ms 0 - 179 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS6490 3.403 ms 0 - 45 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS8550 (Proxy) 1.14 ms 0 - 31 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS8450 1.737 ms 0 - 179 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS9075 1.302 ms 0 - 45 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCM6690 10.44 ms 0 - 236 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 1.495 ms 0 - 122 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.637 ms 0 - 122 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite Mobile 0.728 ms 0 - 126 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS7790 1.495 ms 0 - 122 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS8750 0.728 ms 0 - 126 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS7181 1.117 ms 149 - 149 MB NPU
HRNetPose QNN_DLC float Snapdragon® X2 Elite 1.619 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® X Elite 2.963 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.98 ms 0 - 116 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Gen 1 Mobile 5.016 ms 0 - 102 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8275 14.172 ms 1 - 73 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8550 (Proxy) 2.773 ms 1 - 172 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8450 5.016 ms 0 - 102 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite Mobile 1.567 ms 0 - 73 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA7255P 14.172 ms 1 - 73 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA8295P 4.596 ms 1 - 61 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.278 ms 1 - 76 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS9075 4.118 ms 1 - 3 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8750 1.567 ms 0 - 73 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS7181 2.963 ms 1 - 1 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X2 Elite 1.3 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X Elite 2.091 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.393 ms 0 - 147 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 2.708 ms 0 - 148 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS6490 6.672 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8275 5.164 ms 0 - 99 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8550 (Proxy) 1.881 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8450 2.708 ms 0 - 148 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.823 ms 0 - 101 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.998 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS9075 2.184 ms 2 - 4 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 2.535 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCM6690 19.84 ms 0 - 217 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA8295P 3.065 ms 0 - 98 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA7255P 5.164 ms 0 - 99 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS7790 2.535 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8750 0.998 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS7181 2.091 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X2 Elite 0.682 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X Elite 1.246 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.824 ms 0 - 134 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 1.735 ms 0 - 137 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS6490 3.711 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8275 2.884 ms 0 - 90 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 1.12 ms 0 - 111 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8450 1.735 ms 0 - 137 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS9075 1.273 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA7255P 2.884 ms 0 - 90 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCM6690 10.26 ms 0 - 213 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 1.508 ms 0 - 89 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA8295P 1.964 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.568 ms 0 - 95 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 0.641 ms 0 - 95 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS7790 1.508 ms 0 - 89 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8750 0.641 ms 0 - 95 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS7181 1.246 ms 0 - 0 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Gen 3 Mobile 1.98 ms 0 - 180 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Gen 1 Mobile 5.02 ms 0 - 174 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8275 14.139 ms 0 - 112 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8550 (Proxy) 2.73 ms 0 - 2 MB NPU
HRNetPose TFLITE float Qualcomm® SA8775P 55.36 ms 1 - 34 MB GPU
HRNetPose TFLITE float Qualcomm® SA8650P 55.36 ms 1 - 34 MB GPU
HRNetPose TFLITE float Qualcomm® SA8255P 55.36 ms 1 - 34 MB GPU
HRNetPose TFLITE float Qualcomm® QCS8450 5.02 ms 0 - 174 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite Mobile 1.545 ms 0 - 112 MB NPU
HRNetPose TFLITE float Qualcomm® SA7255P 14.139 ms 0 - 112 MB NPU
HRNetPose TFLITE float Qualcomm® SA8295P 4.646 ms 0 - 103 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 1.284 ms 0 - 116 MB NPU
HRNetPose TFLITE float Qualcomm® QCS9075 4.128 ms 0 - 58 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8750 1.545 ms 0 - 112 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.713 ms 0 - 143 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 1.505 ms 0 - 142 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS6490 3.471 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8275 2.6 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 0.951 ms 0 - 3 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8775P 52.651 ms 1 - 44 MB GPU
HRNetPose TFLITE w8a8 Qualcomm® SA8650P 52.651 ms 1 - 44 MB GPU
HRNetPose TFLITE w8a8 Qualcomm® SA8255P 52.651 ms 1 - 44 MB GPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8450 1.505 ms 0 - 142 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS9075 1.08 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA7255P 2.6 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCM6690 9.542 ms 0 - 204 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 1.359 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8295P 1.78 ms 0 - 85 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.509 ms 0 - 91 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite Mobile 0.58 ms 0 - 89 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS7790 1.359 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8750 0.58 ms 0 - 89 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