{"ID":2833482,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.03745","arxiv_id":"2512.03745","title":"Dual-level Modality Debiasing Learning for Unsupervised Visible-Infrared Person Re-Identification","abstract":"Two-stage learning pipeline has achieved promising results in unsupervised visible-infrared person re-identification (USL-VI-ReID). It first performs single-modality learning and then operates cross-modality learning to tackle the modality discrepancy. Although promising, this pipeline inevitably introduces modality bias: modality-specific cues learned in the single-modality training naturally propagate into the following cross-modality learning, impairing identity discrimination and generalization. To address this issue, we propose a Dual-level Modality Debiasing Learning (DMDL) framework that implements debiasing at both the model and optimization levels. At the model level, we propose a Causality-inspired Adjustment Intervention (CAI) module that replaces likelihood-based modeling with causal modeling, preventing modality-induced spurious patterns from being introduced, leading to a low-biased model. At the optimization level, a Collaborative Bias-free Training (CBT) strategy is introduced to interrupt the propagation of modality bias across data, labels, and features by integrating modality-specific augmentation, label refinement, and feature alignment. Extensive experiments on benchmark datasets demonstrate that DMDL could enable modality-invariant feature learning and a more generalized model. The code is available at https://github.com/priester3/DMDL.","short_abstract":"Two-stage learning pipeline has achieved promising results in unsupervised visible-infrared person re-identification (USL-VI-ReID). It first performs single-modality learning and then operates cross-modality learning to tackle the modality discrepancy. Although promising, this pipeline inevitably introduces modality bi...","url_abs":"https://arxiv.org/abs/2512.03745","url_pdf":"https://arxiv.org/pdf/2512.03745v2","authors":"[\"Jiaze Li\",\"Yan Lu\",\"Bin Liu\",\"Guojun Yin\",\"Mang Ye\"]","published":"2025-12-03T12:43:16Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":606329,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2833482,"paper_url":"https://arxiv.org/abs/2512.03745","paper_title":"Dual-level Modality Debiasing Learning for Unsupervised Visible-Infrared Person Re-Identification","repo_url":"https://github.com/priester3/DMDL","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}