Research map / Detection and segmentation / Open-set and few-shot detection
Object-Generalized Re-Identification: A Step Towards Universal Instance Perception
Shuoyi Chen, Yurui Wu, Mang Ye
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 EDPS and MOIL to the TransReID ViT baseline (ablation) | TransReID ViT baseline · CUTE | mAP: 23.3 -> 31.6 | Improved |
| add EDPS and MOIL to the baseline (ablation) | TransReID ViT baseline · PetFace | mAP: 31.5 -> 40.5 | Improved |
| add EDPS and MOIL to the baseline (ablation) | TransReID ViT baseline · MSMT17 | mAP: 14.2 -> 16.2 | Improved |
| add EDPS and MOIL to the baseline (ablation) | TransReID ViT baseline · Wildlife71 | mAP: 83.3 -> 85.3 | Improved |
| add Hierarchical Semantic Equilibrium on top of EDPS and MOIL (ablation) | MGOR · CUTE | mAP: 31.6 -> 32.9 | Improved |
| add HSE on top of EDPS and MOIL (ablation) | MGOR · MSMT17 | mAP: 16.2 -> 18.1 | Improved |
| add HSE on top of EDPS and MOIL (ablation) | MGOR · PetFace | mAP: 40.5 -> 41.4 | Improved |
| add HSE on top of EDPS and MOIL (ablation) | MGOR · Wildlife71 | mAP: 85.3 -> 86.0 | Improved |
| add HSE on top of EDPS and MOIL (ablation) | MGOR · CUTE | Rank-1: 88.0 -> 87.9 | No clear effect |
| increase category-disjointness between meta-train and meta-test in EDPS (ablation) | MGOR EDPS · MSMT17 vs multi-category sets | mAP | Mixed |
| replace PK sampling with EDPS (ablation) | MGOR · multi-category and single-category sets | mAP | Improved |
| domain-generalization style normalization methods trained on mixed categories (ReNorm, BAU, ADSR) | DG-ReID baselines · MSMT17, UAV-VeID | mAP: BAU 2.7, ADSR 1.2 on MSMT17 mAP | Hurt |
Closest papers by meaning
- Generalizable Object Re-Identification via Visual In-Context Prompting (ICCV 2025 · similarity 0.90)
- Person30K: A Dual-Meta Generalization Network for Person Re-Identification (CVPR 2021 · similarity 0.89)
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 2024 · similarity 0.87)
- Exploring the Camera Bias of Person Re-identification (ICLR 2025 · similarity 0.87)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.87)
- MDReID: Modality-Decoupled Learning for Any-to-Any Multi-Modal Object Re-Identification (NeurIPS 2025 · similarity 0.86)
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.
