Research map / Label-efficient and robust learning / Recognition and long-tail
Modality Unifying Network for Visible-Infrared Person Re-Identification
Hao Yu, Xu Cheng, Wei Peng, Weihao Liu, Guoying Zhao
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 auxiliary modality generated by IML and CML (ablation) | ResNet-50 baseline · SYSU-MM01 all-search | rank-1 (%): 57.49 -> 62.55 | Improved |
| add identity consistency loss L_idc (ablation) | baseline + Aux. · SYSU-MM01 all-search | rank-1 (%): 62.55 -> 66.58 | Improved |
| add identity alignment loss L_ia (ablation) | baseline + Aux. + L_idc · SYSU-MM01 all-search | rank-1 (%): 66.58 -> 71.35 | Improved |
| add modality alignment loss L_ma (ablation) | baseline + Aux. + L_idc · SYSU-MM01 all-search | rank-1 (%): 66.58 -> 69.77 | Improved |
| add both L_ia and L_ma (ablation) | baseline + Aux. + L_idc · SYSU-MM01 all-search | rank-1 (%): 66.58 -> 76.24 | Improved |
| use the proposed auxiliary modality instead of syncretic modality (ablation) | MUN · SYSU-MM01 all-search | rank-1 (%): 72.95 -> 76.24 | Improved |
| use the proposed auxiliary modality instead of X-modality (ablation) | MUN · RegDB | rank-1 (%): 79.95 -> 95.19 | Improved |
| add modality prototype scheme in L_ma (ablation) | modality alignment loss · SYSU-MM01 all-search | rank-1 (%): 69.02 -> 74.15 | Improved |
| add auxiliary bridge scheme in L_ma (ablation) | modality alignment loss · SYSU-MM01 all-search | rank-1 (%): 69.02 -> 71.66 | Improved |
| add full MUN components to baseline under test-time corruption (ablation) | ResNet-50 baseline · SYSU-MM01-C | rank-1 (%): 25.92 -> 41.17 | Improved |
| use MUN instead of CMT on RegDB infrared to visible | VI-ReID method · RegDB infrared to visible | rank-1 (%): 91.97 (CMT) vs 91.86 (MUN) | Hurt |
Closest papers by meaning
- MRCN: A Novel Modality Restitution and Compensation Network for Visible-Infrared Person Re-identification (AAAI 2023 · similarity 0.95)
- Modality-Aware Bias Mitigation and Invariance Learning for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2026 · similarity 0.95)
- FMCNet: Feature-Level Modality Compensation for Visible-Infrared Person Re-Identification (CVPR 2022 · similarity 0.95)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.94)
- Empowering Visible-Infrared Person Re-Identification with Large Foundation Models (NeurIPS 2024 · similarity 0.93)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.92)
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.
