Mutual Information Guided Optimal Transport For Unsupervised Visible-infrared Person Re-identification
2024 · Zhizhong Zhang, Jiangming Wang, Xin Tan, et al.
Abstract
Unsupervised visible infrared person re-identification (USVI-ReID) is a challenging retrieval task that aims to retrieve cross-modality pedestrian images without using any label information. In this task, the large cross-modality variance makes it difficult to generate reliable cross-modality labels, and the lack of annotations also provides additional difficulties for learning modality-invariant features. In this paper, we first deduce an optimization objective for unsupervised VI-ReID based on the mutual information between the model's cross-modality input and output. With equivalent derivation, three learning principles, i.e., "Sharpness" (entropy minimization), "Fairness" (uniform label distribution), and "Fitness" (reliable cross-modality matching) are obtained. Under their guidance, we design a loop iterative training strategy alternating between model training and cross-modality matching. In the matching stage, a uniform prior guided optimal transport assignment ("Fitness", "Fai
Authors
(none)
Tags
Stats
Related papers
- Bridging The Gap: Multi-level Cross-modality Joint Alignment For Visible-infrared Person Re-identification (2023)11.29
- Implicit Discriminative Knowledge Learning For Visible-infrared Person Re-identification (2024)16.19
- Unified Batch All Triplet Loss For Visible-infrared Person Re-identification (2021)9.03
- Multi-memory Matching For Unsupervised Visible-infrared Person Re-identification (2024)12.02
- Video-based Visible-infrared Person Re-identification With Auxiliary Samples (2023)13.49
- Mix-modality Person Re-identification: A New And Practical Paradigm (2024)6.34
- Dynamic Dual-attentive Aggregation Learning For Visible-infrared Person Re-identification (2020)19.67
- BIT: Matching-based Bi-directional Interaction Transformation Network For Visible-infrared Person Re-identification (2026)0.00