Research map / Label-efficient and robust learning / Recognition and long-tail
FMCNet: Feature-Level Modality Compensation for Visible-Infrared Person Re-Identification
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use feature-level modality compensation rather than image-level compensation | FMCNet · SYSU-MM01 and RegDB | Rank-1 | Helped |
| Add SFD to the base feature extractors (ablation) | FMCNet · SYSU-MM01 | Rank-1: 57.09 -> 63.16 | Helped |
| Add FMC to SFD and use compensated modality-specific features (ablation) | FMCNet · SYSU-MM01 | Rank-1: 63.16 -> 65.50 | Helped |
| Add SFF to SFD and FMC (ablation) | FMCNet · SYSU-MM01 | Rank-1: 65.50 -> 66.34 | Helped |
| Add modality-shared feature separation loss Lshs (ablation) | FMCNet MD-loss study · SYSU-MM01 | Rank-1: 58.68 -> 63.96 | Helped |
| Add modality-specific feature separation loss Lsps (ablation) | FMCNet MD-loss study · SYSU-MM01 | Rank-1: 63.96 -> 64.68 | Helped |
| Add decorrelation loss Ldc (ablation) | FMCNet MD-loss study · SYSU-MM01 | Rank-1: 64.68 -> 66.34 | Helped |
| Remove FMC (ablation) | FMCNet · SYSU-MM01 | Rank-1: 66.34 -> 63.16 | Hurt |
| Train FMC with identity and feature consistency losses but no adversarial loss (ablation) | FMCNet · SYSU-MM01 | Rank-1: 57.78 -> 66.34 | Hurt |
| Use adversarial loss in FMC (ablation) | FMCNet · SYSU-MM01 | Rank-1: 57.78 -> 62.56 | Helped |
| Add identity consistency loss LIC to adversarial FMC (ablation) | FMCNet · SYSU-MM01 | Rank-1: 62.56 -> 65.61 | Helped |
| Add feature consistency loss LFC to adversarial FMC with LIC (ablation) | FMCNet · SYSU-MM01 | Rank-1: 65.61 -> 66.34 | Helped |
Closest papers by meaning
- MRCN: A Novel Modality Restitution and Compensation Network for Visible-Infrared Person Re-identification (AAAI 2023 · similarity 0.95)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.92)
- Empowering Visible-Infrared Person Re-Identification with Large Foundation Models (NeurIPS 2024 · similarity 0.91)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.90)
- High-Order Structure Based Middle-Feature Learning for Visible-Infrared Person Re-identification (AAAI 2024 · similarity 0.89)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · 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.