Abstract:Unsupervised person re-identification (ReID), including fully unsupervised ReID and unsupervised domain adaptive ReID, remains a challenge for the fields of biometrics and computer vision due to its difficulty in learning with unlabeled target domain data. Existing state-of-the-art methods, most of which generate pseudo-labels via unsupervised clustering for model optimization, are inevitably hampered by the under-explored problem of pseudo-label noise. Motivated by this, we propose a novel joint framework termed pseudo-label Noise Prevention, Suppression, and Softening (NPSS) for unsupervised person re-identification. Instead of refining generated label noise after clustering as many existing methods do, we start solving this issue from the source of pseudo-label noise by proposing a new Dynamic Camera-Adaptive Clustering (DCAC), which dynamically involves camera information to prevent noise caused by cross-camera variance, thus improving their quality during clustering. Moreover, we propose an Online Domain Union (ODU) mechanism for the classification model learning on the target domain via involving source domain data with their ground-truth labels, which effectively suppresses the indelible noisy pseudo-labels. Furthermore, we present the Self-Consistency Constraint (SCC) to soften the label noise in a single model with reduced computation and network parameter cost, which achieves intra-sample knowledge ensembling with our global-local SCC and cross-sample knowledge ensembling with our inter-instance SCC. Experiments demonstrate the effectiveness of our method as it surpasses state-of-the-art methods by a large margin on Market-1501, DukeMTMC-ReID, and MSMT17 benchmarks. The code is available at https://github.com/hjwang-824/NPSS.

Intensifying the Consistency of Pseudo Label Refinement for Unsupervised Domain Adaptation Person Re-Identification.

Refining pseudo labels for unsupervised Domain Adaptive Re-Identification

Consistency-aware unsupervised label learning for cross-domain person re-identification

Refining Pseudo Labels for Unsupervised Domain Adaptive Person Re-Identification

Federated Domain Adaptation via Pseudo-label Refinement

Soft pseudo-Label shrinkage for unsupervised domain adaptive person re-identification

Domain-adaptive person re-identification via domain alignment and mutual pseudo-label refinement

Progressive Unsupervised Domain Adaptation for Image-based Person Re-Identification

SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification

Delving into Probabilistic Uncertainty for Unsupervised Domain Adaptive Person Re-Identification

Dual-Refinement: Joint Label and Feature Refinement for Unsupervised Domain Adaptive Person Re-Identification

Group-aware Label Transfer for Domain Adaptive Person Re-identification

Unsupervised Person Re-Identification Based on High-Quality Pseudo Labels

Exploiting Reliable Pseudo-Labels for Unsupervised Domain Adaptive Person Re-Identification

Complementary Pseudo Labels for Unsupervised Domain Adaptation on Person Re-identification.

Pseudo-Label Noise Prevention, Suppression and Softening for Unsupervised Person Re-Identification.

Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification

Online Pseudo Label Generation by Hierarchical Cluster Dynamics for Adaptive Person Re-identification

Pseudo Labels Refinement with Intra-camera Similarity for Unsupervised Person Re-identification

Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification

Pseudo Labels Refinement with Stable Cluster Reconstruction for Unsupervised Re-identification.