Research map / Image generation / Super-resolution
Prediction and Recovery for Adaptive Low-Resolution Person Re-Identification
Ke Han, Yan Huang, Zerui Chen, Liang Wang, Tieniu Tan
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 |
|---|---|---|---|
| use adaptive scale factor predictor instead of fixed x2 (ablation) | PRI · MLR-CUHK03 | Rank 1: 82.3 -> 85.2 | Improved |
| use adaptive scale factor predictor instead of fixed x1 (ablation) | PRI · MLR-CUHK03 | Rank 1: 78.6 -> 85.2 | Improved |
| use adaptive predictor instead of size-based predictor (ablation) | PRI · MLR-CUHK03 | Rank 1: 82.5 -> 85.2 | Improved |
| use ideal predictor instead of learned adaptive predictor (ablation) | PRI · MLR-CUHK03 | Rank 1: 85.2 -> 86.8 | Improved |
| use dynamic soft label instead of one-hot label (ablation) | PRI · MLR-CUHK03 | Rank 1: 81.8 -> 85.2 | Improved |
| update soft labels every epoch (omega=1) instead of random update (omega=0) (ablation) | PRI · MLR-CUHK03 and MLR-DukeMTMC-reid | Rank 1 | Improved |
| increase label update interval omega above 1 (ablation) | PRI · MLR-CUHK03 and MLR-DukeMTMC-reid | Rank 1 | Hurt |
| use regulatory factor gamma=1 instead of 0 (ablation) | PRI · MLR-CUHK03 | Rank 1 | Improved |
| add pre-training before joint training (ablation) | PRI · MLR-CUHK03 | Rank 1: 82.8 -> 85.2 | Improved |
| add pre-training before joint training (ablation) | PRI · MLR-DukeMTMC-reid | Rank 1: 75.1 -> 78.3 | Improved |
| add identity loss Lid (ablation) | PRI · MLR-CUHK03 | Rank 1: 80.4 -> 85.2 | Improved |
| add SR loss Lsr (ablation) | PRI · MLR-CUHK03 | Rank 1: 81.5 -> 85.2 | Improved |
Closest papers by meaning
- Deep Low-Resolution Person Re-Identification (AAAI 2018 · similarity 0.90)
- Learning Resolution-Invariant Deep Representations for Person Re-Identification (AAAI 2019 · similarity 0.89)
- Real-World Person Re-Identification via Degradation Invariance Learning (CVPR 2020 · similarity 0.89)
- Recover and Identify: A Generative Dual Model for Cross-Resolution Person Re-Identification (ICCV 2019 · similarity 0.88)
- Resolution as a Direction: Vector-Panning Feature Alignment for Cross-Resolution Re-Identification (ICML 2026 · similarity 0.88)
- Inter-Task Association Critic for Cross-Resolution Person Re-Identification (CVPR 2020 · 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.
