Confidence-guided Centroids For Unsupervised Person Re-identification

Abstract

Unsupervised person re-identification (ReID) aims to train a feature extractor for identity retrieval without exploiting identity labels. Due to the blind trust in imperfect clustering results, the learning is inevitably misled by unreliable pseudo labels. Albeit the pseudo label refinement has been investigated by previous works, they generally leverage auxiliary information such as camera IDs and body part predictions. This work explores the internal characteristics of clusters to refine pseudo labels. To this end, Confidence-Guided Centroids (CGC) are proposed to provide reliable cluster-wise prototypes for feature learning. Since samples with high confidence are exclusively involved in the formation of centroids, the identity information of low-confidence samples, i.e., boundary samples, are NOT likely to contribute to the corresponding centroid. Given the new centroids, current learning scheme, where samples are enforced to learn from their assigned centroids solely, is unwise. To

Stats

• citations8
• S2 citations—
• github stars0
• HF likes0
• heat score7.16
• arxiv keymiao2022confidence

Related papers

• Neighbour Consistency Guided Pseudo-label Refinement For Unsupervised Person Re-identification (2022)0.00
• Part-based Pseudo Label Refinement For Unsupervised Person Re-identification (2022)21.91
• Person Re-identification Via Generalized Class Prototypes (2025)0.00
• Rethinking Person Re-identification From A Projection-on-prototypes Perspective (2023)0.00
• Deep Constrained Dominant Sets For Person Re-identification (2019)9.23
• Fully Unsupervised Person Re-identification Viaselective Contrastive Learning (2020)9.92
• A Semi-supervised Maximum Margin Metric Learning Approach For Small Scale Person Re-identification (2019)4.52
• Prototype-guided Cross-modal Completion And Alignment For Incomplete Text-based Person Re-identification (2023)6.77