limengran98 / SCAE

The code of the paper SCAE: Structural Contrastive Auto-encoder for Incomplete Multi-view Representation Learning

SCAE

The code of the paper SCAE: Structural Contrastive Auto-encoder for Incomplete Multi-view Representation Learning

This repository builds upon the work of TPAMI2022-DCP. We extend our gratitude to all the authors of this work. The datasets can be found here.

Environment Dependencies

To ensure smooth execution, please install the following dependencies:

• pytorch>=1.2.0
• numpy>=1.19.1
• scikit-learn>=0.23.2
• munkres>=1.1.4
• PyGCL>=0.1.0

Usage

To run the multi-view clustering script, use:

python run_clustering_multiview.py

To run the supervised multi-view script, use:

python run_supervised_multiview.py

To run the human action recognition tasks, use:

python run_HAR.py

Citation

If you find our work useful, please cite the following articles:

@article{liscae,
  title={SCAE: Structural Contrastive Auto-encoder for Incomplete Multi-view Representation Learning},
  author={Li, Mengran and Zhang, Ronghui and Zhang, Yong and Piao, Xinglin and Zhao, Shiyu and Yin, Baocai},
  journal={ACM Transactions on Multimedia Computing, Communications and Applications},
  year={2024},
  doi={10.1145/3672078}
}

@article{lin2022dual,
  title={Dual contrastive prediction for incomplete multi-view representation learning},
  author={Lin, Yijie and Gou, Yuanbiao and Liu, Xiaotian and Bai, Jinfeng and Lv, Jiancheng and Peng, Xi},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  volume={45},
  number={4},
  pages={4447--4461},
  year={2022},
  publisher={IEEE}
}