Bottom-up whole-body pose estimation method in constant time.
Official PyTroch implementation of HPRNet.
HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation,
Nermin Samet, Emre Akbas,
- Published at IMAVIS. (arXiv pre-print)*
Model | Body AP | Foot AP | Face AP | Hand AP | Whole-body AP | Download |
---|---|---|---|---|---|---|
HPRNet (DLA) | 55.2 / 57.1 | 49.1 / 50.7 | 74.6 / 75.4 | 47.0 / 48.4 | 31.5 / 32.7 | model |
HPRNet (Hourglass) | 59.4 / 61.1 | 53.0 / 53.9 | 75.4 / 76.0 | 50.4 / 51.2 | 34.8 / 34.9 | model |
train2017
and evaluated on val2017
.[Optional but recommended] create a new conda environment.
conda create --name HPRNet python=3.7
And activate the environment.
conda activate HPRNet
Clone the repo:
HPRNet_ROOT=/path/to/clone/HPRNet
git clone https://github.com/nerminsamet/HPRNet $HPRNet_ROOT
Install PyTorch 1.4.0:
conda install pytorch torchvision cudatoolkit=10.0 -c pytorch
Install the requirements:
pip install -r requirements.txt
First clone the DCNv2 repository into $HPRNet_ROOT/src/lib/models/networks. Then, compile DCNv2 (Deformable Convolutional Networks):
cd $HPRNet_ROOT/src/lib/models/networks/DCNv2
./make.sh
Download the images (2017 Train, 2017 Val) from coco website.
Download train and val annotation files.
${COCO_PATH}
|-- annotations
|-- coco_wholebody_train_v1.0.json
|-- coco_wholebody_val_v1.0.json
|-- images
|-- train2017
|-- val2017
HPRNet_ROOT/models/
.--resume
to resume training.The numerical calculations reported in this paper were fully performed at TUBITAK ULAKBIM, High Performance and Grid Computing Center (TRUBA resources).
HPRNet is released under the MIT License (refer to the LICENSE file for details).
If you find HPRNet useful for your research, please cite our paper as follows:
N. Samet, E. Akbas, "HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation", arXiv, 2021.
BibTeX entry:
@misc{hprnet,
title={HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation},
author={Nermin Samet and Emre Akbas},
year={2021},
}