Research map / Video / Video segmentation and tracking
Region-Based Quality Estimation Network for Large-Scale Person Re-Identification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use region-wise learned quality scores to aggregate frame features | RQEN · iLIDS-VID | CMC Rank-1 accuracy: 64.8 -> 77.1±3.7 | Helped |
| use region-wise learned quality scores to aggregate frame features | RQEN · PRID2011 | CMC Rank-1 accuracy: 81.6 -> 91.8±2.8 | Helped |
| use RQEN rather than the strongest listed baseline, QAN | video person re-identification · PRID 2011 | CMC Rank-1 accuracy: 90.3 -> 91.8±2.8 | Helped |
| use RQEN rather than the strongest listed baseline, QAN | video person re-identification · iLIDS-VID | CMC Rank-1 accuracy: 68.0 -> 77.1±3.7 | Helped |
| use RQEN without XQDA or reranking rather than TriNet+Reranking | person re-identification · MARS | Rank-1 matching rate: 83.03 -> 73.74 | Hurt |
| add upper-region features (ablation) | GoogLeNet+BN baseline · iLIDS-VID | CMC1: 64.8 -> 58.2 | Hurt |
| add middle-region features (ablation) | GoogLeNet+BN baseline · iLIDS-VID | CMC1: 64.8 -> 69.1 | Helped |
| add lower-region features (ablation) | GoogLeNet+BN baseline · iLIDS-VID | CMC1: 64.8 -> 54.9 | Hurt |
| fix regional quality scores to 1 instead of learning them (ablation) | RQEN · iLIDS-VID | CMC1: 72.6±4.8 -> 77.1±3.7 | Hurt |
| replace quality-weighted aggregation with feature concatenation (ablation) | RQEN · iLIDS-VID | CMC1: 43.6 -> 77.1±3.7 | Hurt |
| add upper-region features (ablation) | GoogLeNet+BN baseline · PRID2011 | CMC1: 81.6 -> 76.9 | Hurt |
| add middle-region features (ablation) | GoogLeNet+BN baseline · PRID2011 | CMC1: 81.6 -> 86.1 | Helped |
Closest papers by meaning
- Person Re-Identification in the Wild (CVPR 2017 · similarity 0.89)
- Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning (CVPR 2018 · similarity 0.87)
- Spatial and Temporal Mutual Promotion for Video-Based Person Re-Identification (AAAI 2019 · similarity 0.87)
- Tracklet Self-Supervised Learning for Unsupervised Person Re-Identification (AAAI 2020 · similarity 0.86)
- Large-Scale Pre-Training for Person Re-Identification With Noisy Labels (CVPR 2022 · similarity 0.86)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (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.