Home

Awesome

iCAN (incremental CAN)

Collaborative and Adversarial Network for Unsupervised domain adaptation (CVPR 2018 Spotlight)

Paper: http://openaccess.thecvf.com/content_cvpr_2018/CameraReady/1410.pdf

The code is written with Pytorch 0.2.0. The code with higher version is also available. (e.g. pytorch 0.4.0 for a2w_v2.py and ican_v2.py).

Please Cite:

@inproceedings{zhang2018collaborative,
  title={Collaborative and Adversarial Network for Unsupervised domain adaptation},
  author={Zhang, Weichen and Ouyang, Wanli and Li, Wen and Xu, Dong},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={3801-3809},
  year={2018}
}

We also have a T-PAMI Journal Extension version for CAN which is called Self-Paced Collaborative and Adversarial Network.

Please refer to the following repo for more details.

https://github.com/zhangweichen2006/SPCAN