Domain Generalization数据集整理

一、引言

自从DomainBed在ICLR2021上横空出世之后，Domain Generalization任务看起来好像有了相对统一的标准，后续大量论文都是基于DomainBed框架实现。由于DG的benchmarks非常多，本文将列举主要的几个数据集的下载方式，以及注意事项。

二、Datasets

1、PACS

Download Link
Ps: 打开上述链接后直接点击图片可进入google云盘下载

@inproceedings{Li2017dg,
  			title={Deeper, Broader and Artier Domain Generalization},
  			author={Li, Da and Yang, Yongxin and Song, Yi-Zhe and Hospedales, Timothy},
  			booktitle={International Conference on Computer Vision},
  			year={2017}
		}

2、 Office-Home

Download Link
Ps: 打开上述链接后找到路径可进入google云盘下载

@inproceedings{venkateswara2017deep,
  title={Deep hashing network for unsupervised domain adaptation},
  author={Venkateswara, Hemanth and Eusebio, Jose and Chakraborty, Shayok and Panchanathan, Sethuraman},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={5018--5027},
  year={2017}
}

3、 VLCS

Download Link
Ps: 打开上述链接后找到路径可进入google云盘下载

@INPROCEEDINGS{6751316,
  author={Fang, Chen and Xu, Ye and Rockmore, Daniel N.},
  booktitle={2013 IEEE International Conference on Computer Vision}, 
  title={Unbiased Metric Learning: On the Utilization of Multiple Datasets and Web Images for Softening Bias}, 
  year={2013},
  volume={},
  number={},
  pages={1657-1664},
  doi={10.1109/ICCV.2013.208}}

4、TerraIncognita （subset）

Download Link
Ps: 标签提取比较麻烦

@inproceedings{beery2018recognition,
  title={Recognition in terra incognita},
  author={Beery, Sara and Van Horn, Grant and Perona, Pietro},
  booktitle={Proceedings of the European conference on computer vision (ECCV)},
  pages={456--473},
  year={2018}
}

5、DomainNet

Download Link

@inproceedings{peng2019moment,
  title={Moment matching for multi-source domain adaptation},
  author={Peng, Xingchao and Bai, Qinxun and Xia, Xide and Huang, Zijun and Saenko, Kate and Wang, Bo},
  booktitle={Proceedings of the IEEE International Conference on Computer Vision},
  pages={1406--1415},
  year={2019}
}