Research map / Label-efficient and robust learning / Recognition and long-tail
RGB-Infrared Cross-Modality Person Re-Identification
Ancong Wu, Wei-Shi Zheng, Hong-Xing Yu, Shaogang Gong, Jianhuang Lai
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 deep zero-padding to the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 12.04 -> 14.80 | Improved |
| add deep zero-padding to the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 all-search single-shot | mAP (%): 13.67 -> 15.95 | Improved |
| add deep zero-padding to the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 all-search multi-shot | rank-1 accuracy (%): 16.26 -> 19.13 | Improved |
| add deep zero-padding to the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 indoor-search single-shot | rank-1 accuracy (%): 16.94 -> 20.58 | Improved |
| add deep zero-padding to the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 indoor-search multi-shot | rank-1 accuracy (%): 22.62 -> 24.43 | Improved |
| replace two-stream network with one-stream deep zero-padding | RGB-IR Re-ID network · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 11.65 -> 14.80 | Improved |
| replace asymmetric FC layer network with one-stream deep zero-padding | RGB-IR Re-ID network · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 9.30 -> 14.80 | Improved |
| use explicit domain-indicator channels with the one-stream network (ablation) | one-stream ResNet-6 · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 12.04 -> 11.79 | No clear effect |
| use deep zero-padding instead of domain indicator channels (ablation) | one-stream ResNet-6 · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 11.79 -> 14.80 | Improved |
| use deep network instead of Lin's GSM two-stream | RGB-IR Re-ID network · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 5.29 -> 11.65 | Improved |
| use deep zero-padding instead of handcrafted features with metric learning | RGB-IR Re-ID network · SYSU-MM01 all-search single-shot | rank-1 accuracy (%): 3.64 -> 14.80 | Improved |
Closest papers by meaning
- Cross-Modality Paired-Images Generation for RGB-Infrared Person Re-Identification (AAAI 2020 · similarity 0.94)
- Learning Modal-Invariant and Temporal-Memory for Video-Based Visible-Infrared Person Re-Identification (CVPR 2022 · similarity 0.93)
- RGB-Infrared Cross-Modality Person Re-Identification via Joint Pixel and Feature Alignment (ICCV 2019 · similarity 0.93)
- Robust Multi-Modality Person Re-identification (AAAI 2021 · similarity 0.91)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.89)
- Cross-Modality Perturbation Synergy Attack for Person Re-identification (NeurIPS 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.
