Research map / Vision-language and multimodal / Human and embodied modeling
Image-Text Knowledge Modeling for Unsupervised Multi-Scenario Person Re-Identification
Zhiqi Pang, Lingling Zhao, Yang Liu, Chunyu Wang, Gaurav Sharma
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 scenario embedding (ablation) | M1 (PGM baseline) · SYSU-MM01 (All Search) | Rank-1: 53.8 -> 56.1 | Improved |
| add scenario embedding (ablation) | M1 (PGM baseline) · LTCC (Clothing Change) | Rank-1: 15.2 -> 17.6 | Improved |
| add scenario embedding (ablation) | M1 (PGM baseline) · MLR-CUHK03 | Rank-1: 33.2 -> 34.4 | Improved |
| add multi-scenario separation loss (ablation) | M2 · SYSU-MM01 (All Search) | Rank-1: 56.1 -> 57.0 | Improved |
| add multi-scenario separation loss (ablation) | M2 · MLR-CUHK03 | Rank-1: 34.4 -> 36.3 | Improved |
| add dynamic text representation update (DRU) (ablation) | M3 · SYSU-MM01 (All Search) | Rank-1: 57.0 -> 59.3 | Improved |
| add dynamic text representation update (DRU) (ablation) | M3 · LTCC (Clothing Change) | Rank-1: 18.5 -> 20.2 | Improved |
| replace PGM with cluster-level heterogeneous matching (CHM) (ablation) | M4 · MLR-CUHK03 | Rank-1: 37.0 -> 50.7 | Improved |
| replace PGM with cluster-level heterogeneous matching (CHM) (ablation) | M4 · SYSU-MM01 (All Search) | Rank-1: 59.3 -> 62.6 | Improved |
| add instance-level heterogeneous matching (IHM) (ablation) | M5 · MLR-CUHK03 | Rank-1: 50.7 -> 63.6 | Improved |
| use IHM instead of CNL for instance-level positive sets (ablation) | M6 · LTCC (Clothing Change) | Rank-1: 25.6 -> 27.3 | Improved |
| train SDCL jointly across three scenarios instead of single scenario | SDCL · SYSU-MM01 (All Search) | Rank-1: 64.5 -> 63.0 | Hurt |
Closest papers by meaning
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.89)
- Unifying Multi-Modal Uncertainty Modeling and Semantic Alignment for Text-to-Image Person Re-identification (AAAI 2024 · similarity 0.89)
- Modeling Thousands of Human Annotators for Generalizable Text-to-Image Person Re-identification (CVPR 2025 · similarity 0.88)
- Multi-Prompts Learning with Cross-Modal Alignment for Attribute-Based Person Re-identification (AAAI 2024 · similarity 0.88)
- Human-centered Interactive Learning via MLLMs for Text-to-Image Person Re-identification (CVPR 2025 · similarity 0.87)
- LLaVA-ReID: Selective Multi-image Questioner for Interactive Person Re-Identification (ICML 2025 · similarity 0.87)
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.
