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 1.24.1 Download
ONNX w8a16 Universal QAIRT 2.42, ONNX Runtime 1.24.1 Download
ONNX w8a8 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
QNN_DLC w8a8 Universal QAIRT 2.43 Download
TFLITE float Universal QAIRT 2.43, TFLite 2.17.0 Download
TFLITE w8a8 Universal QAIRT 2.43, TFLite 2.17.0 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 2.382 ms 55 - 55 MB NPU
HRNetPose ONNX float Snapdragon® X Elite 2.62 ms 55 - 55 MB NPU
HRNetPose ONNX float Snapdragon® 8 Gen 3 Mobile 1.86 ms 0 - 145 MB NPU
HRNetPose ONNX float Qualcomm® QCS8550 (Proxy) 2.432 ms 0 - 58 MB NPU
HRNetPose ONNX float Qualcomm® QCS9075 3.904 ms 0 - 4 MB NPU
HRNetPose ONNX float Snapdragon® 8 Elite For Galaxy Mobile 1.503 ms 0 - 93 MB NPU
HRNetPose ONNX float Snapdragon® 8 Elite Gen 5 Mobile 1.217 ms 1 - 91 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X2 Elite 1.012 ms 28 - 28 MB NPU
HRNetPose ONNX w8a16 Snapdragon® X Elite 1.985 ms 28 - 28 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 1.278 ms 0 - 202 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS6490 476.311 ms 29 - 33 MB CPU
HRNetPose ONNX w8a16 Qualcomm® QCS8550 (Proxy) 1.721 ms 0 - 37 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCS9075 2.0 ms 0 - 3 MB NPU
HRNetPose ONNX w8a16 Qualcomm® QCM6690 222.648 ms 30 - 49 MB CPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.976 ms 0 - 131 MB NPU
HRNetPose ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 216.257 ms 26 - 40 MB CPU
HRNetPose ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.787 ms 0 - 128 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X2 Elite 0.781 ms 30 - 30 MB NPU
HRNetPose ONNX w8a8 Snapdragon® X Elite 1.647 ms 28 - 28 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 1.05 ms 0 - 186 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS6490 91.319 ms 9 - 68 MB CPU
HRNetPose ONNX w8a8 Qualcomm® QCS8550 (Proxy) 1.476 ms 0 - 33 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCS9075 1.542 ms 0 - 3 MB NPU
HRNetPose ONNX w8a8 Qualcomm® QCM6690 60.963 ms 9 - 35 MB CPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.927 ms 0 - 124 MB NPU
HRNetPose ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 58.834 ms 9 - 30 MB CPU
HRNetPose ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.808 ms 0 - 122 MB NPU
HRNetPose QNN_DLC float Snapdragon® X2 Elite 1.638 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® X Elite 2.964 ms 1 - 1 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.971 ms 0 - 119 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8275 (Proxy) 14.16 ms 1 - 75 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8550 (Proxy) 2.735 ms 1 - 2 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA8775P 4.311 ms 0 - 76 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS9075 4.136 ms 3 - 5 MB NPU
HRNetPose QNN_DLC float Qualcomm® QCS8450 (Proxy) 4.963 ms 0 - 106 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA7255P 14.16 ms 1 - 75 MB NPU
HRNetPose QNN_DLC float Qualcomm® SA8295P 4.571 ms 1 - 68 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 1.55 ms 0 - 75 MB NPU
HRNetPose QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.245 ms 1 - 78 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X2 Elite 1.281 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® X Elite 2.156 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.391 ms 0 - 150 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS6490 6.791 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8275 (Proxy) 5.146 ms 0 - 98 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8550 (Proxy) 1.89 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA8775P 2.283 ms 0 - 101 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS9075 2.202 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCM6690 20.146 ms 0 - 217 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® QCS8450 (Proxy) 2.614 ms 0 - 150 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA7255P 5.146 ms 0 - 98 MB NPU
HRNetPose QNN_DLC w8a16 Qualcomm® SA8295P 3.142 ms 0 - 97 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 1.028 ms 0 - 99 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 2.538 ms 0 - 103 MB NPU
HRNetPose QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.797 ms 0 - 100 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X2 Elite 0.7 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® X Elite 1.288 ms 0 - 0 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.841 ms 0 - 133 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS6490 3.782 ms 2 - 4 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8275 (Proxy) 2.836 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 1.157 ms 0 - 4 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA8775P 1.505 ms 0 - 90 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS9075 1.32 ms 0 - 2 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCM6690 10.307 ms 0 - 212 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® QCS8450 (Proxy) 1.678 ms 0 - 136 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA7255P 2.836 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Qualcomm® SA8295P 1.982 ms 0 - 87 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.662 ms 0 - 92 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 1.515 ms 0 - 88 MB NPU
HRNetPose QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.568 ms 0 - 93 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Gen 3 Mobile 1.975 ms 0 - 191 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8275 (Proxy) 14.187 ms 0 - 114 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8550 (Proxy) 2.727 ms 0 - 3 MB NPU
HRNetPose TFLITE float Qualcomm® SA8775P 18.722 ms 0 - 116 MB NPU
HRNetPose TFLITE float Qualcomm® QCS9075 4.156 ms 0 - 58 MB NPU
HRNetPose TFLITE float Qualcomm® QCS8450 (Proxy) 4.945 ms 0 - 177 MB NPU
HRNetPose TFLITE float Qualcomm® SA7255P 14.187 ms 0 - 114 MB NPU
HRNetPose TFLITE float Qualcomm® SA8295P 4.603 ms 0 - 105 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 1.558 ms 0 - 116 MB NPU
HRNetPose TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 1.278 ms 0 - 118 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.73 ms 0 - 143 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS6490 3.379 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8275 (Proxy) 2.564 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 0.955 ms 0 - 2 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8775P 1.374 ms 0 - 89 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS9075 1.11 ms 0 - 30 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCM6690 9.588 ms 0 - 208 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® QCS8450 (Proxy) 1.499 ms 0 - 142 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA7255P 2.564 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Qualcomm® SA8295P 1.79 ms 0 - 85 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.582 ms 0 - 88 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 1.356 ms 0 - 86 MB NPU
HRNetPose TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.517 ms 0 - 91 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