Papers › Style Normalization and Restitution for Generalizable Person Re-identification

Style Normalization and Restitution for Generalizable Person Re-identification

22 May 2020CVPR 2020 6arXiv:2005.11037archive 2025-07-28

Xin Jin, Cuiling Lan, Wen-Jun Zeng, Zhibo Chen, Li Zhang

Existing fully-supervised person re-identification (ReID) methods usually suffer from poor generalization capability caused by domain gaps. The key to solving this problem lies in filtering out identity-irrelevant interference and learning domain-invariant person representations. In this paper, we aim to design a generalizable person ReID framework which trains a model on source domains yet is able to generalize/perform well on target domains. To achieve this goal, we propose a simple yet effective Style Normalization and Restitution (SNR) module. Specifically, we filter out style variations (e.g., illumination, color contrast) by Instance Normalization (IN). However, such a process inevitably removes discriminative information. We propose to distill identity-relevant feature from the removed information and restitute it to the network to ensure high discrimination. For better disentanglement, we enforce a dual causal loss constraint in SNR to encourage the separation of identity-relevant features and identity-irrelevant features. Extensive experiments demonstrate the strong generalization capability of our framework. Our models empowered by the SNR modules significantly outperform the state-of-the-art domain generalization approaches on multiple widely-used person ReID benchmarks, and also show superiority on unsupervised domain adaptation.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

microsoft/SNR officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DisentanglementDomain AdaptationDomain GeneralizationGeneralizable Person Re-identificationPerson Re-IdentificationUnsupervised Domain Adaptation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Unsupervised Domain Adaptation CUHK03 to MSMT SNR R1 22.0 #7 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT SNR R10 - #7 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT SNR R5 - #7 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT SNR mAP 7.7 #7 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market SNR R1 77.8 #9 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market SNR R10 - #9 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market SNR R5 - #9 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market SNR mAP 52.4 #9 of 9 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market SNR mAP 61.7 #11 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market SNR rank-1 82.8 #11 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market SNR rank-10 - #11 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market SNR rank-5 - #11 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 SNR R1 17.1 #7 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 SNR R10 - #7 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 SNR R5 - #7 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 SNR mAP 17.5 #7 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke SNR mAP 58.1 #9 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke SNR rank-1 76.3 #9 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke SNR rank-10 - #9 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke SNR rank-5 - #9 of 25 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Instance Normalization

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections