Research map / Label-efficient and robust learning / Recognition and long-tail
OIMNet++: Prototypical Normalization and Localization-Aware Learning for Person Search
Sanghoon Lee, Youngmin Oh, Donghyeon Baek, Junghyup Lee, Bumsub Ham
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 BatchNorm before L2 normalization (ablation) | OIMNet (OIM loss) · PRW search | mAP: 42.0 -> 44.3 | Improved |
| replace BatchNorm with ProtoNorm (ablation) | OIMNet (OIM loss) · PRW search | mAP: 44.3 -> 46.3 | Improved |
| add ProtoNorm without BatchNorm (ablation) | OIMNet (OIM loss) · PRW search | mAP: 42.0 -> 46.3 | Improved |
| replace BatchNorm with ProtoNorm (ablation) | OIMNet (OIM loss) · PRW reID | mAP: 46.6 -> 48.4 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with BatchNorm · PRW search | mAP: 44.3 -> 45.1 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with BatchNorm · PRW search | rank-1: 81.5 -> 82.9 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with ProtoNorm · PRW search | mAP: 46.3 -> 46.8 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with ProtoNorm · PRW search | rank-1: 82.7 -> 83.9 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with ProtoNorm · PRW reID | rank-1: 84.6 -> 86.2 | Improved |
| replace OIM loss with LOIM loss (ablation) | OIMNet with BatchNorm · PRW reID | mAP: 46.6 -> 47.4 | Improved |
| combine ProtoNorm and LOIM loss (ablation) | OIMNet · PRW search | mAP: 42.0 -> 46.8 | Improved |
| replace BatchNorm with ProtoNorm (ablation) | OIMNet with LOIM loss · PRW reID | mAP: 47.4 -> 49.0 | Improved |
Later papers that compared with this method
2 later accepted papers report OIMNet+ as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Context-aware Dynamic Contrastive Learning Network and E-Bike Rider Benchmark for Person Search (AAAI 2026 · better on 2 of 2 shared results)
- Domain Generalizable Person Search Using Unreal Dataset (AAAI 2024 · better on 3 of 3 shared results)
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
- Norm-Aware Embedding for Efficient Person Search (CVPR 2020 · similarity 0.89)
- Joint Detection and Identification Feature Learning for Person Search (CVPR 2017 · similarity 0.88)
- Diverse Knowledge Distillation for End-to-End Person Search (AAAI 2021 · similarity 0.88)
- Efficient Person Search: An Anchor-Free Approach (IJCV 2023 · similarity 0.88)
- Divide and Conquer: Hybrid Pre-training for Person Search (AAAI 2024 · similarity 0.87)
- Cheb-GR: Rethinking K-nearest Neighbor Search in Re-ranking for Person Re-identification (CVPR 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.
