Research map / Label-efficient and robust learning / Recognition and long-tail
Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning
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 EUG with progressive pseudo-label sampling instead of the one-shot baseline | ETAP-Net · MARS | rank-1: 36.16 -> 62.67 | Helped |
| Use EUG with progressive pseudo-label sampling instead of Stepwise | ETAP-Net · MARS | rank-1: 41.21 -> 62.67 | Helped |
| Use EUG with progressive pseudo-label sampling instead of Stepwise | ETAP-Net · DukeMTMC-VideoReID | rank-1: 56.26 -> 72.79 | Helped |
| Lower the enlarging factor from p = 0.30 to p = 0.05 (ablation) | EUG · MARS | rank-1: 42.77 -> 62.67 | Helped |
| Lower the enlarging factor from p = 0.30 to p = 0.05 (ablation) | EUG · DukeMTMC-VideoReID | rank-1: 63.82 -> 72.79 | Helped |
| Select pseudo-labels by feature-space dissimilarity rather than classification loss (ablation) | EUG · MARS | rank-1: 48.33 -> 62.67 | Helped |
| Select pseudo-labels by feature-space dissimilarity rather than classification loss (ablation) | EUG (p = 0.05) · MARS | mAP: 25.35 -> 42.45 | Helped |
| Use a larger enlarging factor, p = 0.30 instead of p = 0.05 (ablation) | EUG · MARS | rank-1: 62.67 -> 42.77 | Hurt |
| Use a larger enlarging factor, p = 0.30 instead of p = 0.05 (ablation) | EUG · DukeMTMC-VideoReID | mAP: 63.23 -> 54.57 | Hurt |
| Increase the selected pseudo-labeled subset progressively rather than selecting a fixed large quantity from the beginning | EUG · MARS | rank-1: 41.21 -> 62.67 | Helped |
Closest papers by meaning
- Tracklet Self-Supervised Learning for Unsupervised Person Re-Identification (AAAI 2020 · similarity 0.92)
- Large-Scale Pre-Training for Person Re-Identification With Noisy Labels (CVPR 2022 · similarity 0.90)
- Keypoint Message Passing for Video-Based Person Re-identification (AAAI 2022 · similarity 0.89)
- Intra-Inter Camera Similarity for Unsupervised Person Re-Identification (CVPR 2021 · similarity 0.89)
- Person Re-Identification in the Wild (CVPR 2017 · similarity 0.89)
- Unsupervised Pre-Training for Person Re-Identification (CVPR 2021 · 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.