Hugging Face's logo

qualcomm
/
HRNetPose

Keypoint Detection
PyTorch
LiteRT
ONNX
android
Model card Files Community

HRNetPose: Optimized for Mobile Deployment

Perform accurate human pose estimation

HRNet performs pose estimation in high-resolution representations.

This model is an implementation of HRNetPose found here.

This repository provides scripts to run HRNetPose on Qualcomm® devices. More details on model performance across various devices, can be found here.

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 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
HRNetPose float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 102.912 ms 0 - 71 MB NPU HRNetPose.tflite
HRNetPose float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 14.216 ms 1 - 10 MB NPU Use Export Script
HRNetPose float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 3.926 ms 0 - 114 MB NPU HRNetPose.tflite
HRNetPose float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 5.128 ms 0 - 51 MB NPU Use Export Script
HRNetPose float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 2.666 ms 0 - 185 MB NPU HRNetPose.tflite
HRNetPose float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 2.678 ms 1 - 3 MB NPU Use Export Script
HRNetPose float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 4.453 ms 0 - 71 MB NPU HRNetPose.tflite
HRNetPose float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 4.339 ms 1 - 14 MB NPU Use Export Script
HRNetPose float SA7255P ADP Qualcomm® SA7255P TFLITE 102.912 ms 0 - 71 MB NPU HRNetPose.tflite
HRNetPose float SA7255P ADP Qualcomm® SA7255P QNN 14.216 ms 1 - 10 MB NPU Use Export Script
HRNetPose float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 2.674 ms 0 - 34 MB NPU HRNetPose.tflite
HRNetPose float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 2.646 ms 1 - 2 MB NPU Use Export Script
HRNetPose float SA8295P ADP Qualcomm® SA8295P TFLITE 4.557 ms 0 - 67 MB NPU HRNetPose.tflite
HRNetPose float SA8295P ADP Qualcomm® SA8295P QNN 4.558 ms 1 - 18 MB NPU Use Export Script
HRNetPose float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 2.666 ms 0 - 81 MB NPU HRNetPose.tflite
HRNetPose float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 2.689 ms 1 - 3 MB NPU Use Export Script
HRNetPose float SA8775P ADP Qualcomm® SA8775P TFLITE 4.453 ms 0 - 71 MB NPU HRNetPose.tflite
HRNetPose float SA8775P ADP Qualcomm® SA8775P QNN 4.339 ms 1 - 14 MB NPU Use Export Script
HRNetPose float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 2.668 ms 0 - 19 MB NPU HRNetPose.tflite
HRNetPose float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 2.636 ms 1 - 14 MB NPU Use Export Script
HRNetPose float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 2.743 ms 0 - 134 MB NPU HRNetPose.onnx
HRNetPose float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 1.956 ms 0 - 114 MB NPU HRNetPose.tflite
HRNetPose float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 2.004 ms 1 - 54 MB NPU Use Export Script
HRNetPose float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 2.11 ms 0 - 92 MB NPU HRNetPose.onnx
HRNetPose float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 1.853 ms 0 - 75 MB NPU HRNetPose.tflite
HRNetPose float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 1.56 ms 1 - 40 MB NPU Use Export Script
HRNetPose float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 1.739 ms 0 - 52 MB NPU HRNetPose.onnx
HRNetPose float Snapdragon X Elite CRD Snapdragon® X Elite QNN 2.847 ms 1 - 1 MB NPU Use Export Script
HRNetPose float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 2.694 ms 55 - 55 MB NPU HRNetPose.onnx
HRNetPose w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 2.632 ms 0 - 48 MB NPU HRNetPose.tflite
HRNetPose w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 2.882 ms 0 - 10 MB NPU Use Export Script
HRNetPose w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 1.365 ms 0 - 96 MB NPU HRNetPose.tflite
HRNetPose w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 1.871 ms 0 - 83 MB NPU Use Export Script
HRNetPose w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 0.994 ms 0 - 132 MB NPU HRNetPose.tflite
HRNetPose w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 1.139 ms 0 - 4 MB NPU Use Export Script
HRNetPose w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 1.327 ms 0 - 50 MB NPU HRNetPose.tflite
HRNetPose w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 1.419 ms 0 - 14 MB NPU Use Export Script
HRNetPose w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) TFLITE 3.843 ms 0 - 74 MB NPU HRNetPose.tflite
HRNetPose w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) QNN 5.222 ms 0 - 11 MB NPU Use Export Script
HRNetPose w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 17.399 ms 0 - 2 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 2.632 ms 0 - 48 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA7255P ADP Qualcomm® SA7255P QNN 2.882 ms 0 - 10 MB NPU Use Export Script
HRNetPose w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 0.984 ms 0 - 131 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 1.154 ms 0 - 3 MB NPU Use Export Script
HRNetPose w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 1.734 ms 0 - 51 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA8295P ADP Qualcomm® SA8295P QNN 1.879 ms 0 - 18 MB NPU Use Export Script
HRNetPose w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 0.984 ms 0 - 135 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 1.138 ms 0 - 2 MB NPU Use Export Script
HRNetPose w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 1.327 ms 0 - 50 MB NPU HRNetPose.tflite
HRNetPose w8a8 SA8775P ADP Qualcomm® SA8775P QNN 1.419 ms 0 - 14 MB NPU Use Export Script
HRNetPose w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 0.956 ms 0 - 133 MB NPU HRNetPose.tflite
HRNetPose w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 1.162 ms 0 - 18 MB NPU Use Export Script
HRNetPose w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 5.021 ms 3 - 103 MB NPU HRNetPose.onnx
HRNetPose w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 0.724 ms 0 - 88 MB NPU HRNetPose.tflite
HRNetPose w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 0.852 ms 0 - 80 MB NPU Use Export Script
HRNetPose w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 3.4 ms 0 - 164 MB NPU HRNetPose.onnx
HRNetPose w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 0.694 ms 0 - 53 MB NPU HRNetPose.tflite
HRNetPose w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 0.735 ms 0 - 52 MB NPU Use Export Script
HRNetPose w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 3.238 ms 0 - 122 MB NPU HRNetPose.onnx
HRNetPose w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN 1.261 ms 0 - 0 MB NPU Use Export Script
HRNetPose w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 5.833 ms 26 - 26 MB NPU HRNetPose.onnx

Installation

Install the package via pip:

pip install "qai-hub-models[hrnet-pose]" torch==2.4.1 -f https://download.openmmlab.com/mmcv/dist/cpu/torch2.4/index.html -f https://qaihub-public-python-wheels.s3.us-west-2.amazonaws.com/index.html

Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.hrnet_pose.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.hrnet_pose.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

  • Performance check on-device on a cloud-hosted device
  • Downloads compiled assets that can be deployed on-device for Android.
  • Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.hrnet_pose.export

Profiling Results
------------------------------------------------------------
HRNetPose
Device                          : cs_8275 (ANDROID 14)                 
Runtime                         : TFLITE                               
Estimated inference time (ms)   : 102.9                                
Estimated peak memory usage (MB): [0, 71]                              
Total # Ops                     : 516                                  
Compute Unit(s)                 : npu (516 ops) gpu (0 ops) cpu (0 ops)

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.hrnet_pose import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S24")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.hrnet_pose.demo --on-device

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.hrnet_pose.demo -- --on-device

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

  • TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.

  • QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on HRNetPose's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of HRNetPose can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month
151
Inference Providers NEW
Keypoint Detection
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support