Research map / Label-efficient and robust learning / Recognition and long-tail
Exploring Homogeneous and Heterogeneous Consistent Label Associations for Unsupervised Visible-Infrared Person ReID
Lingfeng He, De Cheng, Nannan Wang, Xinbo Gao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add cross-modality loss L_CM (MULT) to the baseline (ablation) | Baseline · SYSU-MM01 all-search | mAP / Rank-1: +16.35% mAP, +18.26% Rank-1 | Improved |
| train within entire AMIRL framework with MULT (ablation) | Baseline · SYSU-MM01 all-search | mAP / Rank-1: +20.66% mAP, +23.01% Rank-1 | Improved |
| add OCLR directly to the baseline (ablation) | Baseline · SYSU-MM01 all-search | mAP / Rank-1: +12.46% mAP, +12.58% Rank-1 | Improved |
| add OCLR to AMIRL framework (ablation) | Baseline + L_IM + L_CM · SYSU-MM01 all-search | mAP / Rank-1: +3.71% mAP, +4.59% Rank-1 | Improved |
| add AMIRL (intra-modality memory structure) without OCLR (ablation) | Baseline + L_CM · SYSU-MM01 all-search | mAP: +0.60% mAP | Improved |
| add AMIRL with OCLR (ablation) | Baseline + L_CM + L_OCLR · SYSU-MM01 all-search | mAP: +2.41% mAP | Improved |
| use both intra and auxiliary memory instead of intra only (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 63.77 -> 64.77 | Improved |
| use both intra and auxiliary memory instead of auxiliary only (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 63.06 -> 64.77 | Improved |
| use AMIRL instead of MIRL (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 64.05 -> 64.77 | Improved |
| model heterogeneous relation S^he with optimal transport instead of Jaccard (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 61.08 -> 64.77 | Improved |
| model S^he with optimal transport instead of cosine (k=20) (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 62.19 -> 64.77 | Improved |
| add channel augmentation (CA) to OCLR (ablation) | Ours · SYSU-MM01 all-search | Rank-1: 61.98 -> 64.77 | Improved |
Closest papers by meaning
- Towards Grand Unified Representation Learning for Unsupervised Visible-Infrared Person Re-Identification (ICCV 2023 · similarity 0.94)
- Relieving Universal Label Noise for Unsupervised Visible-Infrared Person Re-Identification by Inferring from Neighbors (AAAI 2025 · similarity 0.93)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.91)
- Empowering Visible-Infrared Person Re-Identification with Large Foundation Models (NeurIPS 2024 · similarity 0.90)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.90)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
