Abstract:Unsupervised domain adaptive person re-identification aims at adapting the re-identification model trained on a labeled source domain to an unlabeled target domain. The mainstream pipeline alternates between clustering-based pseudo-label prediction and representation learning, but the imperfect interaction between these steps generates noisy pseudo labels that diminish the model’s effectiveness. Previous methods reduce noisy pseudo labels impact by assessing consistency only at a single granularity level, overlooking multi-level confidence analysis for better feature representation. To address the issue, we propose a novel Multi-Granularity Consistency Network (MGCN) to perform the noise-resistance learning across different granularity consistency perspectives, including prototype-wise consistency, triplet-wise consistency and list-wise consistency, to suppress the contribution of noisy samples simultaneously. Specifically, the prototype-wise consistency leverages the prototypical output affinity between teacher and student networks to evaluate the reliability of pseudo label of a target sample, thus reducing their negative impact on identity classification loss. Triplet-wise consistency focuses on the triplet distance discrepancies between the teacher and student networks to retain the reliable and informative samples that satisfy triplet distance constraint in the triplet loss, thereby facilitating more effective model training and improved performance in the target domain. Furthermore, the list-wise consistency uses accurate list-wise similarity rankings from the teacher’s memory bank to select more dependable neighboring samples in the student’s memory bank, pulling these closer in feature space to alleviate the detrimental effects of noisy labels in contrastive loss. Based on the multi-granularity consistency, MGCN evaluates the credibility of pseudo labels and adjusts their impact across three re-ID losses for effective domain adaptation. Experimental results demonstrate that our proposed method achieves significant improvements over the existing methods on multiple benchmarks.

Learning with noisy labels method for unsupervised domain adaptive person re-identification

Refining Pseudo Labels for Unsupervised Domain Adaptive Person Re-Identification

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

NPLP: A Noisy Pseudo-Label Processing Approach for Unsupervised Domain-Adaptive Person Re-ID

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

Noise Resistible Network for Unsupervised Domain Adaptation on Person Re-Identification

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

Dual-Uncertainty Guided Curriculum Learning and Part-Aware Feature Refinement for Domain Adaptive Person Re-Identification

Unsupervised Domain Adaptive Person Re-Identification Via Camera Penalty Learning.

Large-Scale Pre-training for Person Re-identification with Noisy Labels

Unsupervised cross-domain person re-identification with self-attention and joint-flexible optimization

Refining pseudo labels for unsupervised Domain Adaptive Re-Identification

Noise-Resistance Learning Via Multi-Granularity Consistency for Unsupervised Domain Adaptive Person Re-Identification

ANL: Anti-Noise Learning for Cross-Domain Person Re-Identification

Adapt Only Once: Fast Unsupervised Person Re-Identification Via Relevance-Aware Guidance

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

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

Relieving Universal Label Noise for Unsupervised Visible-Infrared Person Re-Identification by Inferring from Neighbors

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

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

Refining Noisy Labels with Label Reliability Perception for Person Re-identification