Papers › Hard-sample Guided Hybrid Contrast Learning for Unsupervised Person Re-Identification

Hard-sample Guided Hybrid Contrast Learning for Unsupervised Person Re-Identification

25 Sep 2021arXiv:2109.12333archive 2025-07-28

Zheng Hu, Chuang Zhu, Gang He

Unsupervised person re-identification (Re-ID) is a promising and very challenging research problem in computer vision. Learning robust and discriminative features with unlabeled data is of central importance to Re-ID. Recently, more attention has been paid to unsupervised Re-ID algorithms based on clustered pseudo-label. However, the previous approaches did not fully exploit information of hard samples, simply using cluster centroid or all instances for contrastive learning. In this paper, we propose a Hard-sample Guided Hybrid Contrast Learning (HHCL) approach combining cluster-level loss with instance-level loss for unsupervised person Re-ID. Our approach applies cluster centroid contrastive loss to ensure that the network is updated in a more stable way. Meanwhile, introduction of a hard instance contrastive loss further mines the discriminative information. Extensive experiments on two popular large-scale Re-ID benchmarks demonstrate that our HHCL outperforms previous state-of-the-art methods and significantly improves the performance of unsupervised person Re-ID. The code of our work is available soon at https://github.com/bupt-ai-cz/HHCL-ReID.

PaperPDFCode

Code

bupt-ai-cz/hhcl-reid officialmentioned in papermentioned on GitHubpytorch 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

Contrastive LearningPerson Re-IdentificationPseudo LabelUnsupervised Person Re-Identification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Unsupervised Person Re-Identification DukeMTMC-reID HHCL(ResNet50 w/o RK) MAP 73.3 #3 of 13 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID HHCL(ResNet50 w/o RK) Rank-1 85.1 #3 of 13 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID HHCL(ResNet50 w/o RK) Rank-10 94.6 #3 of 13 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID HHCL(ResNet50 w/o RK) Rank-5 92.4 #3 of 13 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501 HHCL(ResNet50 w/o RK) MAP 84.2 #9 of 23 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501 HHCL(ResNet50 w/o RK) Rank-1 93.4 #9 of 23 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501 HHCL(ResNet50 w/o RK) Rank-10 98.5 #9 of 23 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501 HHCL(ResNet50 w/o RK) Rank-5 97.7 #9 of 23 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.

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