Research map / Vision-language and multimodal / Cross-modal learning
Weakly Supervised Text-Based Person Re-Identification
Shizhen Zhao, Changxin Gao, Yuanjie Shao, Wei-Shi Zheng, Nong Sang
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 MPLR to the baseline (ablation) | baseline · CUHK-PEDES | Rank-1: 51.13 -> 53.28 | Improved |
| add Text-IoU CMPM on top of MPLR (ablation) | baseline + MPLR · CUHK-PEDES | Rank-1: 53.28 -> 55.59 | Improved |
| add Text-IoU HSM on top of MPLR and Text-IoU CMPM (ablation) | CMMT · CUHK-PEDES | Rank-1: 55.59 -> 57.10 | Improved |
| add MPLR v->t refinement (ablation) | CMMT · CUHK-PEDES | Rank-1: 56.03 -> 57.10 | Improved |
| add MPLR t->v refinement (ablation) | CMMT · CUHK-PEDES | Rank-1: 55.96 -> 57.10 | Improved |
| add both MPLR directions (ablation) | CMMT · CUHK-PEDES | Rank-1: 54.92 -> 57.10 | Improved |
| use Text-IoU soft-labels instead of textual-embedding soft-labels in CMPM (ablation) | CMMT · CUHK-PEDES | Rank-1: 50.15 -> 57.10 | Improved |
| use Text-IoU soft-labels instead of visual-embedding soft-labels in CMPM (ablation) | CMMT · CUHK-PEDES | Rank-1: 48.59 -> 57.10 | Improved |
| add Text-IoU CHSM (ablation) | CMMT · CUHK-PEDES | Rank-1: 56.02 -> 57.10 | Improved |
| add Text-IoU IHSM (ablation) | CMMT · CUHK-PEDES | Rank-1: 56.42 -> 57.10 | Improved |
| embedding-based soft-labels instead of baseline in-batch negatives (ablation) | baseline · CUHK-PEDES | Rank-1: 51.13 -> 50.15 | Hurt |
| vary loss weights alpha in 0.6-1.0 and beta in 0.5-1.0 (ablation) | CMMT · CUHK-PEDES | Rank-1 | No clear effect |
Closest papers by meaning
- Human-centered Interactive Learning via MLLMs for Text-to-Image Person Re-identification (CVPR 2025 · similarity 0.87)
- AXM-Net: Implicit Cross-Modal Feature Alignment for Person Re-identification (AAAI 2022 · similarity 0.87)
- Noisy-Correspondence Learning for Text-to-Image Person Re-identification (CVPR 2024 · similarity 0.86)
- Unifying Multi-Modal Uncertainty Modeling and Semantic Alignment for Text-to-Image Person Re-identification (AAAI 2024 · similarity 0.86)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.85)
- Modeling Thousands of Human Annotators for Generalizable Text-to-Image Person Re-identification (CVPR 2025 · similarity 0.85)
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.
