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

TL;DR

This paper introduces a neighbor-guided approach to reduce label noise in unsupervised visible-infrared person re-identification, improving accuracy by replacing hard pseudo labels with soft labels from neighbors and dynamically weighting samples.

Contribution

It proposes the N-ULC and N-DW modules that mitigate universal label noise using neighbor information, enhancing the stability and performance of USL-VI-ReID models.

Findings

Outperforms existing methods on RegDB and SYSU-MM01 datasets.

Effectively reduces label noise with neighbor-guided soft labels.

Improves training stability through dynamic sample weighting.

Abstract

Unsupervised visible-infrared person re-identification (USL-VI-ReID) is of great research and practical significance yet remains challenging due to the absence of annotations. Existing approaches aim to learn modality-invariant representations in an unsupervised setting. However, these methods often encounter label noise within and across modalities due to suboptimal clustering results and considerable modality discrepancies, which impedes effective training. To address these challenges, we propose a straightforward yet effective solution for USL-VI-ReID by mitigating universal label noise using neighbor information. Specifically, we introduce the Neighbor-guided Universal Label Calibration (N-ULC) module, which replaces explicit hard pseudo labels in both homogeneous and heterogeneous spaces with soft labels derived from neighboring samples to reduce label noise. Additionally, we present the Neighbor-guided Dynamic Weighting (N-DW) module to enhance training stability by minimizing the influence of unreliable samples. Extensive experiments on the RegDB and SYSU-MM01 datasets demonstrate that our method outperforms existing USL-VI-ReID approaches, despite its simplicity. The source code is available at: https://github.com/tengxiao14/Neighbor-guided-USL-VI-ReID.