Research map / Label-efficient and robust learning / Recognition and long-tail
Augmented and Softened Matching for Unsupervised Visible-Infrared Person Re-Identification
Zhiqi Pang, Chunyu Wang, Lingling Zhao, Junjie Wang
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 augmented matching (CAM) in place of PGM (ablation) | ASM · SYSU-MM01 All Search | Rank-1: 56.87 -> 59.63 | Improved |
| add CAM in place of PGM (ablation) | ASM · RegDB Visible to Thermal | Rank-1: 68.51 -> 75.80 | Improved |
| replace Laccl with cross-modality soft contrastive loss Lcsc (with PGM) (ablation) | ASM · SYSU-MM01 All Search | Rank-1: 56.87 -> 58.21 | Improved |
| replace Laccl with Lcsc (with CAM) (ablation) | ASM · SYSU-MM01 All Search | Rank-1: 59.63 -> 62.87 | Improved |
| add cross-modality hard contrastive loss Lchc (ablation) | ASM · SYSU-MM01 All Search | Rank-1: 62.87 -> 65.07 | Improved |
| add Lchc on top of CAM + Lcsc (ablation) | ASM · SYSU-MM01 Indoor Search | Rank-1: 68.89 -> 71.08 | Improved |
| soft labels via momentum update stabilize training vs hard-label Laccl | ASM · SYSU-MM01 / RegDB | training loss stability | Improved |
| ASM vs SDCL on SYSU-MM01 All Search | SDCL · SYSU-MM01 All Search | Rank-1: 64.49 -> 65.07 | Improved |
| ASM vs SDCL on SYSU-MM01 Indoor Search | SDCL · SYSU-MM01 Indoor Search | Rank-1: 71.37 -> 71.08 | Hurt |
| ASM vs SDCL on RegDB Visible to Thermal | SDCL · RegDB Visible to Thermal | Rank-1: 86.91 -> 88.23 | Improved |
Closest papers by meaning
- Multi-Memory Matching for Unsupervised Visible-Infrared Person Re-Identification (ECCV 2024 · similarity 0.92)
- Dual-Level Matching With Outlier Filtering for Unsupervised Visible-Infrared Person Re-Identification (TPAMI 2025 · similarity 0.91)
- Unsupervised Visible-Infrared Person Re-Identification via Progressive Graph Matching and Alternate Learning (CVPR 2023 · similarity 0.90)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.90)
- Relieving Universal Label Noise for Unsupervised Visible-Infrared Person Re-Identification by Inferring from Neighbors (AAAI 2025 · similarity 0.89)
- Learning Commonality, Divergence and Variety for Unsupervised Visible-Infrared Person Re-identification (NeurIPS 2024 · similarity 0.88)
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.
