Home

Awesome

EventHPE: Event-based 3D Human Pose and Shape Estimation

Shihao Zou, Chuan Guo, Xinxin Zuo, Sen Wang, Xiaoqin Hu, Shoushun Chen, Minglun Gong and Li Cheng. ICCV 2021.

Dataset

You can download the data from Google Drive or Microsoft OneDrive, which consists of

Requirements

python 3.7.5
torch 1.7.0
opendr 0.78 (for render SMPL shape, installed successfully only under ubuntu 18.04)
cv2 4.1.1

To download the SMPL model go to this project website and register to get access to the downloads section. Place under ./smpl_model. The model version used in our project is

basicModel_m_lbs_10_207_0_v1.0.0.pkl
basicModel_neutral_lbs_10_207_0_v1.0.0.pkl

Citation

If you would like to use our code or dataset, please cite either

@inproceedings{zou2021eventhpe,  
  title={EventHPE: Event-based 3D Human Pose and Shape Estimation},  
  author={Zou, Shihao and Guo, Chuan and Zuo, Xinxin and Wang, Sen and Xiaoqin, Hu and Chen, Shoushun and Gong, Minglun and Cheng, Li},  
  booktitle={Proceedings of the IEEE International Conference on Computer Vision},  
  year={2021}  
} 

Our following work

Event-based Human Pose Tracking using Spiking Spatiotemporal Transformer [project]