Research map / Label-efficient and robust learning / Recognition and long-tail
Learning by Aligning: Visible-Infrared Person Re-Identification Using Cross-Modal Correspondences
Hyunjong Park, Sanghoon Lee, Junghyup Lee, Bumsub Ham
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 ID consistency loss with CMAlign (ablation) | ID-loss baseline · SYSU-MM01 all-search (ablation) | mAP (%): 49.54 -> 52.88 | Improved |
| add dense triplet loss with co-attention map and CMAlign (ablation) | ID-loss baseline · SYSU-MM01 all-search (ablation) | mAP (%): 49.54 -> 51.23 | Improved |
| add dense triplet loss without co-attention map (ablation) | ID-loss baseline · SYSU-MM01 all-search (ablation) | mAP (%): 49.54 -> 50.08 | No clear effect |
| add co-attention map to dense triplet loss (with ID consistency) (ablation) | L_IC + L_DT · SYSU-MM01 all-search (ablation) | mAP (%): 52.78 -> 54.14 | Improved |
| add co-attention map to dense triplet loss (with ID consistency) (ablation) | L_IC + L_DT · SYSU-MM01 all-search (ablation) | rank-1 (%): 53.44 -> 55.41 | Improved |
| add dense triplet loss with co-attention to ID consistency loss (ablation) | L_IC · SYSU-MM01 all-search (ablation) | mAP (%): 52.88 -> 54.14 | Improved |
| add ID consistency loss to dense triplet loss with co-attention (ablation) | L_DT + A · SYSU-MM01 all-search (ablation) | mAP (%): 51.23 -> 54.14 | Improved |
| add CMAlign at both conv4-6 and conv5-3 instead of conv4-6 only (ablation) | full model · SYSU-MM01 all-search (ablation) | mAP (%): 53.02 -> 54.14 | Improved |
| add CMAlign at both conv4-6 and conv5-3 instead of conv5-3 only (ablation) | full model · SYSU-MM01 all-search (ablation) | mAP (%): 53.81 -> 54.14 | Improved |
| use ours instead of DDAG on indoor-search | Ours · SYSU-MM01 indoor-search | mAP (%): 67.98 -> 66.33 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Hi-CMD — Hi-CMD: Hierarchical Cross-Modality Disentanglement for Visible-Infrared Person Re-Identification (CVPR 2020) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Hi-CMD: Hierarchical Cross-Modality Disentanglement for Visible-Infrared Person Re-Identification (CVPR 2020 · similarity 0.93)
- Cross-Modality Person Re-identification with Memory-Based Contrastive Embedding (AAAI 2023 · similarity 0.93)
- High-Order Structure Based Middle-Feature Learning for Visible-Infrared Person Re-identification (AAAI 2024 · similarity 0.92)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.92)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.91)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.91)
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.
