Research map / Label-efficient and robust learning / Recognition and long-tail
One-Shot Metric Learning for Person Re-Identification
Check 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 |
|---|---|---|---|
| add a learned one-shot color metric with spatial background-distortion coefficients (ablation) | one-shot re-identification method · VIPeR, CUHK01, iLIDS, PRID | rank-1 accuracy: 30.1 -> 34.3; 39.6 -> 45.6; 49.9 -> 51.2; 31.9 -> 41.4 | Helped |
| use intensity-only training for the deep texture representation instead of color-image training (ablation) | JSTL on unseen datasets · PRID | rank-1 accuracy: 2.0 -> 21.0 | Helped |
| use intensity-only training for the deep texture representation instead of color-image training (ablation) | JSTL on unseen datasets · iLIDS | rank-1 accuracy: 44.1 -> 44.0 | No clear effect |
| combine deep intensity texture features with the one-shot color metric rather than use either alone (ablation) | one-shot re-identification method · VIPeR, CUHK01, iLIDS, PRID | rank-1 accuracy | Helped |
| allow patches to move horizontally during assignment (ablation) | one-shot re-identification method · VIPeR | rank-1 accuracy: +3.2% | Helped |
| allow patches to move vertically during assignment (ablation) | one-shot re-identification method · all evaluated datasets | rank-1 accuracy | Hurt |
| learn a color metric from ColorChecker patches rather than use standard color calibration methods | color dissimilarity on CCH · CCH | rank-1 accuracy: 94.57% OUR,α=α(n); 91.09% OUR,α=1; 81.74% OUR,α=0 | Helped |
| use spatially learned background-distortion coefficients rather than a fixed α = 0 (ablation) | color metric on CCH · CCH | rank-1 accuracy: 81.74% -> 94.57% | Helped |
Closest papers by meaning
- Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning (CVPR 2018 · similarity 0.88)
- One-Shot Image Classification by Learning to Restore Prototypes (AAAI 2020 · similarity 0.87)
- See the Forest for the Trees: Joint Spatial and Temporal Recurrent Neural Networks for Video-Based Person Re-Identification (CVPR 2017 · similarity 0.87)
- Resource Aware Person Re-Identification Across Multiple Resolutions (CVPR 2018 · similarity 0.87)
- Easy Identification From Better Constraints: Multi-Shot Person Re-Identification From Reference Constraints (CVPR 2018 · similarity 0.85)
- Single Camera Training for Person Re-Identification (AAAI 2020 · 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.