Part-based Pseudo Label Refinement For Unsupervised Person Re-identification

Abstract

Unsupervised person re-identification (re-ID) aims at learning discriminative representations for person retrieval from unlabeled data. Recent techniques accomplish this task by using pseudo-labels, but these labels are inherently noisy and deteriorate the accuracy. To overcome this problem, several pseudo-label refinement methods have been proposed, but they neglect the fine-grained local context essential for person re-ID. In this paper, we propose a novel Part-based Pseudo Label Refinement (PPLR) framework that reduces the label noise by employing the complementary relationship between global and part features. Specifically, we design a cross agreement score as the similarity of k-nearest neighbors between feature spaces to exploit the reliable complementary relationship. Based on the cross agreement, we refine pseudo-labels of global features by ensembling the predictions of part features, which collectively alleviate the noise in global feature clustering. We further refine pseudo

Stats

• citations261
• S2 citations—
• github stars76
• HF likes0
• heat score21.91
• arxiv keycho2022part

Related papers

• Neighbour Consistency Guided Pseudo-label Refinement For Unsupervised Person Re-identification (2022)0.00
• Confidence-guided Centroids For Unsupervised Person Re-identification (2022)7.16
• Sparse Label Smoothing Regularization For Person Re-identification (2018)9.41
• Body Part-based Representation Learning For Occluded Person Re-identification (2022)21.32
• Prototype-guided Cross-modal Completion And Alignment For Incomplete Text-based Person Re-identification (2023)6.77
• Improving Person Re-identification By Attribute And Identity Learning (2017)20.71
• Deep-person: Learning Discriminative Deep Features For Person Re-identification (2017)16.90
• Multi-memory Matching For Unsupervised Visible-infrared Person Re-identification (2024)12.02