Dual Distribution Alignment Network for Generalizable Person Re-Identification

Chen, Peixian1; Dai, Pingyang1; Liu, Jianzhuang3; Zheng, Feng2; Xu, Mingliang4; Tian, Qi5; Ji, Rongrong1,6

2021

35th AAAI Conference on Artificial Intelligence / 33rd Conference on Innovative Applications of Artificial Intelligence / 11th Symposium on Educational Advances in Artificial Intelligence

2159-5399

2374-3468

THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE

35

1054-1062

FEB 02-09, 2021

null,null,ELECTR NETWORK

2275 E BAYSHORE RD, STE 160, PALO ALTO, CA 94303 USA

ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE

Domain generalization (DG) offers a preferable real-world setting for Person Re-Identification (Re-ID), which trains a model using multiple source domain datasets and expects it to perform well in an unseen target domain without any model updating. Unfortunately, most DG approaches are designed explicitly for classification tasks, which fundamentally differs from the retrieval task Re-ID. Moreover, existing applications of DG in Re-ID cannot correctly handle the massive variation among Re-ID datasets. In this paper, we identify two fundamental challenges in DG for Person Re-ID: domain-wise variations and identity-wise similarities. To this end, we propose an end-to-end Dual Distribution Alignment Network (DDAN) to learn domain-invariant features with dual-level constraints: the domain-wise adversarial feature learning and the identity-wise similarity enhancement. These constraints effectively reduce the domain-shift among multiple source domains further while agreeing to real-world scenarios. We evaluate our method in a large-scale DG Re-ID benchmark and compare it with various cutting-edge DG approaches. Quantitative results show that DDAN achieves state-of-the-art performance.

其他

英语

CPCI-S ; EI

National Science Fund for Distinguished Young[62025603]

Computer Science ; Education & Educational Research

Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications ; Education, Scientific Disciplines

WOS:000680423501016

20222012115628

Web of Science

1.Xiamen Univ, Sch Informat, Dept Artificial Intelligence, Media Analyt & Comp Lab, Xiamen, Peoples R China
2.Southern Univ Sci & Technol, Dept Comp Sci & Engn, Shenzhen, Peoples R China
3.Huawei Tech, Noahs Ark Lab, Shenzhen, Peoples R China
4.Zhengzhou Univ, Sch Informat Engn, Zhengzhou, Henan, Peoples R China
5.Huawei Tech, Cloud & AI, Shenzhen, Peoples R China
6.Xiamen Univ, Inst Artificial Intelligence, Xiamen, Peoples R China

Chen, Peixian,Dai, Pingyang,Liu, Jianzhuang,et al. Dual Distribution Alignment Network for Generalizable Person Re-Identification[C]. 2275 E BAYSHORE RD, STE 160, PALO ALTO, CA 94303 USA:ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE,2021:1054-1062.