Home

Awesome

Discovering Human-Object Interaction Concepts via Self-Compositional Learning

This repository contains the official implementation of the paper "Discovering Human-Object Interaction Concepts via Self-Compositional Learning", which is accepted to ECCV2022.

This is the SCL code based on QPIC. Please also refer to HOI-CL for the code of SCL based on Tensorflow Currently we have provided the code of Concept Discovery with Qpic.

Code is preparing.

Training and Inference

Please refer to QPIC for Training and Inference.

Citation

Please consider citing our paper if it helps your research.

@inproceedings{hou2022scl,
  title={Discovering Human-Object Interaction Concepts via Self-Compositional Learning},
  author={Hou, Zhi and Yu, Baosheng and Tao, Dacheng},
  booktitle={ECCV},
  year={2022}
}

Acknowledgement

Code is based on QPIC