Joint Generative and Contrastive Learning for Unsupervised Person Re-identification

被引:160
作者
Chen, Hao [1 ,2 ,3 ]
Wang, Yaohui [1 ,2 ]
Lagadec, Benoit [3 ]
Dantcheva, Antitza [1 ,2 ]
Bremond, Francois [1 ,2 ]
机构
[1] INRIA, Le Chesnay, France
[2] Univ Cote Azur, Nice, France
[3] European Syst Integrat, Le Cannet, France
来源
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021 | 2021年
关键词
D O I
10.1109/CVPR46437.2021.00204
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recent self-supervised contrastive learning provides an effective approach for unsupervised person re-identification (ReID) by learning invariance from different views (transformed versions) of an input. In this paper, we incorporate a Generative Adversarial Network (GAN) and a contrastive learning module into one joint training framework. While the GAN provides online data augmentation for contrastive learning, the contrastive module learns view-invariant features for generation. In this context, we propose a mesh-based view generator. Specifically, mesh projections serve as references towards generating novel views of a person. In addition, we propose a view-invariant loss to facilitate contrastive learning between original and generated views. Deviating from previous GAN-based unsupervised ReID methods involving domain adaptation, we do not rely on a labeled source dataset, which makes our method more flexible. Extensive experimental results show that our method significantly outperforms state-of-the-art methods under both, fully unsupervised and unsupervised domain adaptive settings on several large scale ReID datsets.
引用
收藏
页码:2004 / 2013
页数:10
相关论文
共 54 条
[1]  
[Anonymous], 2020, ICML
[2]  
[Anonymous], 2017, NEURIPS
[3]  
[Anonymous], 2018, NEURIPS
[4]  
[Anonymous], 2020, CELL REP PHYS SCI
[5]  
[Anonymous], 2020, Computer Vision and Pattern Recognition, DOI DOI 10.1109/CVPR42600.2020.00401
[6]  
[Anonymous], 2020, AAAI
[7]  
[Anonymous], 2018, CVPR, DOI DOI 10.1109/CVPR.2018.00016
[8]   Domain Adaptation Through Synthesis for Unsupervised Person Re-identification [J].
Bak, Slawomir ;
Carr, Peter ;
Lalonde, Jean-Francois .
COMPUTER VISION - ECCV 2018, PT XIII, 2018, 11217 :193-209
[9]  
Brock A., 2019, INT C LEARNING REPRE
[10]   Everybody Dance Now [J].
Chan, Caroline ;
Ginosar, Shiry ;
Zhou, Tinghui ;
Efros, Alexei A. .
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :5932-5941