Research map / Image generation / Super-resolution
Learning Resolution-Invariant Deep Representations for 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 the multi-level rather than single-level discriminator (ablation) | RAIN · MLR-CUHK03 | Rank 1: 77.6 -> 78.9 | Helped |
| Use the multi-level rather than single-level discriminator (ablation) | RAIN · MLR-VIPeR | Rank 1: 41.2 -> 42.5 | Helped |
| Use the multi-level rather than single-level discriminator (ablation) | RAIN · CAVIAR | Rank 1: 41.5 -> 42.0 | Helped |
| Add classification loss (ablation) | RAIN · MLR-CUHK03 | Rank 1: 70.8 -> 78.9 | Helped |
| Add triplet loss (ablation) | RAIN · MLR-CUHK03 | Rank 1: 69.1 -> 78.9 | Helped |
| Add HR reconstruction loss (ablation) | RAIN · MLR-CUHK03 | Rank 1: 67.3 -> 78.9 | Helped |
| Add adversarial loss for cross-resolution feature alignment (ablation) | RAIN · MLR-CUHK03 | Rank 1: 65.9 -> 78.9 | Helped |
| Train with multiple LR down-sampling rates rather than one rate (ablation) | RAIN · MLR-CUHK03 | Rank 1: 77.2 -> 78.9 | Helped |
| Train with multiple LR down-sampling rates rather than one rate (ablation) | RAIN · MLR-VIPeR | Rank 1: 41.5 -> 42.5 | Helped |
| Train with multiple LR down-sampling rates rather than one rate (ablation) | RAIN · CAVIAR | Rank 1: 40.1 -> 42.0 | Helped |
| Apply RAIN to a test resolution not seen during training (r=8) | RAIN · MLR-VIPeR, unseen resolution r=8 | Rank 1: 37.5 | Helped |
| Increase labeled training data from 0% to 20% | RAIN · MLR-CUHK03 semi-supervised | Rank 1: 3.3% -> 66.3% | Helped |
Closest papers by meaning
- Deep Low-Resolution Person Re-Identification (AAAI 2018 · similarity 0.93)
- Inter-Task Association Critic for Cross-Resolution Person Re-Identification (CVPR 2020 · similarity 0.90)
- Resolution as a Direction: Vector-Panning Feature Alignment for Cross-Resolution Re-Identification (ICML 2026 · similarity 0.90)
- Resource Aware Person Re-Identification Across Multiple Resolutions (CVPR 2018 · similarity 0.89)
- Real-World Person Re-Identification via Degradation Invariance Learning (CVPR 2020 · similarity 0.89)
- Deep Neural Networks with Inexact Matching for Person Re-Identification (NeurIPS 2016 · similarity 0.89)
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.