idealwei / ASANet

The code and trained models of: Affinity Space Adaptation for Semantic Segmentation Across Domains.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Affinity Space Adaptation for Semantic Segmentation Across Domains


Pytorch implementation of the paper "Affinity Space Adaptation for Semantic Segmentation Across Domains", TIP, 2020. In this paper, we address the problem of unsupervised domain adaptation (UDA) in semantic segmentation, achieving the state-of-the-art performance on standard benchmarks.

Paper

If you find this paper useful in your research, please consider citing:

@ARTICLE{9184275,
  author={W. {Zhou} and Y.{Wang} and J. {Chu} and J. {Yang} and X. {Bai} and Y. {Xu}},
  journal={IEEE Transactions on Image Processing}, 
  title={Affinity Space Adaptation for Semantic Segmentation Across Domains}, 
  year={2020},
  volume={},
  number={},
  pages={1-1},}

Example Results

Quantitative Reuslts

  1. Comparison Results on Cityscapes when adapted from GTA5 in terms of per-class IoU and mIoU over 19 class.
  2. Comparison Results on Cityscapes when adapted from SYTNTHIA in terms of per-class IoU and mIoU over 13 or 16 class.

Usage

Datasets

Initial Weights

Initial weights and trained models can be downloaded from here. [Google Drive] [Baidu Drive (download code: 9lov) ].
Put the weights in the "ASANet/pretrained" directory.

Training Script:

bash scripts/train_gta2city.sh

Testing Scripts:

bash scripts/evaluate.sh

Contact

Thanks for your attention! If you have any suggestion or question, you can leave a message here or contact us directly:

About

The code and trained models of: Affinity Space Adaptation for Semantic Segmentation Across Domains.


Languages

Language:Python 97.6%Language:Shell 2.4%