Research map / Label-efficient and robust learning / Semi-supervised learning
Semi-Supervised Bayesian Attribute Learning 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 |
|---|---|---|---|
| Learn attributes with IBP prior and re-ID probability jointly (SBAL) vs best prior joint method SSDAL+XQDA | SBAL · VIPeR supervised | Rank-1 accuracy (%): 43.5 -> 45.2 | Helped |
| SBAL vs best prior method on PRID | SBAL · PRID supervised | Rank-1 accuracy (%): 29.80 -> 32.4 | Helped |
| SBAL vs DLLR with 2/3 of training pairs unlabeled | SBAL · VIPeR semi-supervised | Rank-1 accuracy (%): 32.5 -> 33.6 | Helped |
| SBAL vs DLLR with 2/3 of training pairs unlabeled | SBAL · PRID semi-supervised | Rank-1 accuracy (%): 22.1 -> 24.4 | Helped |
| SBAL without predefined attributes vs ACRN using predefined attributes | SBAL · DukeMTMC-reID | CMC Rank-1 (%): 72.58 -> 71.03 | Hurt |
| SBAL vs ACRN on mAP | SBAL · DukeMTMC-reID | mAP (%): 51.96 -> 52.42 | Helped |
| Increase number of latent attributes K+ up to the automatically learned optimum (ablation) | SBAL · VIPeR and PRID | CMC Rank-1 (%) | Helped |
| Force K+ beyond the learned optimum (1740 VIPeR, 1588 PRID) up to 2000 (ablation) | SBAL · VIPeR and PRID | CMC Rank-1 (%) | Hurt |
Closest papers by meaning
- Learning a Discriminative Null Space for Person Re-Identification (CVPR 2016 · similarity 0.90)
- A Bottom-Up Clustering Approach to Unsupervised Person Re-Identification (AAAI 2019 · similarity 0.89)
- Point to Set Similarity Based Deep Feature Learning for Person Re-Identification (CVPR 2017 · similarity 0.89)
- Unsupervised Pre-Training for Person Re-Identification (CVPR 2021 · similarity 0.88)
- Unsupervised Person Re-Identification via Softened Similarity Learning (CVPR 2020 · similarity 0.88)
- CA-Jaccard: Camera-aware Jaccard Distance for Person Re-identification (CVPR 2024 · 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.