Research map / Label-efficient and robust learning / Recognition and long-tail
Instruct-ReID++: Towards Universal Purpose Instruction-Guided Person Re-Identification
Weizhen He, Yiheng Deng, Yunfeng Yan, Feng Zhu, Yizhou Wang, Lei Bai, Qingsong Xie, Rui Zhao, Donglian Qi, Wanli Ouyang, Shixiang Tang
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 editing transformer instruction integration (ablation) | IRM (MTL) · Average over 11 test sets | mAP: 64.5 -> 65.1 | Improved |
| add editing transformer instruction integration (ablation) | IRM (STL) · Average over 11 test sets | mAP: 61.5 -> 61.8 | Improved |
| add adaptive triplet loss instead of triplet loss (ablation) | IRM (MTL) · Average over 11 test sets | mAP: 64.8 -> 65.1 | Improved |
| add adaptive triplet loss on varied-instruction CTCC-ReID (ablation) | IRM (STL) · COCAS+ Real2 (CTCC-ReID) | mAP: 31.5 -> 32.2 | Improved |
| add adaptive triplet loss on varied-instruction LI-ReID (ablation) | IRM (STL) · COCAS+ Real2 (LI-ReID) | mAP: 30.2 -> 30.7 | Improved |
| add multi-task learning | IRM · LTCC | mAP: 46.7 -> 52.0 | Improved |
| add multi-task learning | IRM · VC-Clothes | mAP: 80.1 -> 78.9 | Hurt |
| add multi-task learning | IRM · COCAS+ Real2 (CTCC-ReID) | mAP: 32.2 -> 41.7 | Improved |
| replace triplet loss with memory-bank contrastive learning (IRM++) | IRM (STL) · COCAS+ Real2 (CTCC-ReID) | mAP: 22.2 -> 32.2 | Improved |
| replace triplet loss with memory-bank contrastive learning (IRM++) | IRM (STL) · COCAS+ Real2 (LI-ReID) | mAP: 28.7 -> 35.8 | Improved |
| replace triplet loss with memory-bank contrastive learning (IRM++) | IRM (MTL) · COCAS+ Real2 (CTCC-ReID) | mAP: 26.3 -> 35.3 | Improved |
| replace triplet loss with memory-bank contrastive learning (IRM++) | IRM · Joint test set (task-free) | mAP: 56.2 -> 65.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PASS — PASS: Part-Aware Self-Supervised Pre-training for Person Re-identification (ECCV 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.96)
- Exploring the Camera Bias of Person Re-identification (ICLR 2025 · similarity 0.89)
- LLaVA-ReID: Selective Multi-image Questioner for Interactive Person Re-Identification (ICML 2025 · similarity 0.89)
- Multi-Prompts Learning with Cross-Modal Alignment for Attribute-Based Person Re-identification (AAAI 2024 · similarity 0.89)
- Learning Continual Compatible Representation for Re-indexing Free Lifelong Person Re-identification (CVPR 2024 · similarity 0.89)
- Person De-reidentification: A Variation-guided Identity Shift Modeling (CVPR 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.
