TorchSemiSeg

[CVPR 2021] Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision

by Xiaokang Chen¹, Yuhui Yuan², Gang Zeng¹, Jingdong Wang².

¹ Key Laboratory of Machine Perception (MOE), Peking University ² Microsoft Research Asia.

[Poster] [Video (YouTube)]

Simpler Is Better !

News

• [July 9 2021] We have released some SOTA methods (Mean Teacher, CCT, GCT).
• [June 3 2021] Please check our paper in Arxiv. Data and code have been released.

Installation

Please refer to the Installation document.

Getting Started

Please follow the Getting Started document.

Citation

Please consider citing this project in your publications if it helps your research.

@inproceedings{chen2021-CPS,
  title={Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision},
  author={Chen, Xiaokang and Yuan, Yuhui and Zeng, Gang and Wang, Jingdong},
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2021}
}

TODO

• [x] Dataset release
• [x] Code for CPS + CutMix
• [x] Code for Cityscapes dataset
• [x] Other SOTA semi-supervised segmentation methods