Research map / Label-efficient and robust learning / Recognition and long-tail
Unsupervised Person Re-Identification by Deep Asymmetric Metric Embedding
Hong-Xing Yu, Ancong Wu, Wei‐Shi Zheng
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 asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · CUHK01 (single-shot) | rank-1 (%): 55.95 -> 65.81 | Improved |
| use asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · CUHK03 (single-shot) | rank-1 (%): 27.86 -> 38.27 | Improved |
| use asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · SYSU (single-shot) | rank-1 (%): 25.38 -> 36.14 | Improved |
| use asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · Market-1501 (multi-shot) | rank-1 (%): 49.94 -> 60.24 | Improved |
| use asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · ExMarket (multi-shot) | rank-1 (%): 53.00 -> 62.98 | Improved |
| use asymmetric metric instead of symmetric metric (DECMEL) (ablation) | DECAMEL · MSMT17 (multi-shot) | rank-1 (%): 23.88 -> 30.34 | Improved |
| use asymmetric metric instead of symmetric metric (ablation) | DECAMEL · SYSU (single-shot) | S value (inter/intra): 1.61 -> 1.88 | Improved |
| add DECAMEL asymmetric metric clustering on top of JSTL feature | JSTL feature · Market-1501 (multi-shot) | mAP (%): 18.36 -> 32.44 | Improved |
| add DECAMEL on top of JSTL feature | JSTL feature · CUHK01 (single-shot) | rank-1 (%): 46.26 -> 65.81 | Improved |
| add DECAMEL on top of JSTL feature | JSTL feature · MSMT17 (multi-shot) | rank-1 (%): 21.24 -> 30.34 | Improved |
| use asymmetric metric clustering instead of symmetric clustering model UsNCA | UsNCA · SYSU (single-shot) | rank-1 (%): 21.07 -> 36.14 | Improved |
| set lambda to zero (remove cross-view consistency regularization) (ablation) | DECAMEL · SYSU | matching rate | Hurt |
Closest papers by meaning
- Cross-View Asymmetric Metric Learning for Unsupervised Person Re-Identification (ICCV 2017 · similarity 0.96)
- Person Re-Identification by Camera Correlation Aware Feature Augmentation (TPAMI 2017 · similarity 0.92)
- Asymmetric Co-Teaching for Unsupervised Cross-Domain Person Re-Identification (AAAI 2020 · similarity 0.91)
- Stepwise Metric Promotion for Unsupervised Video Person Re-Identification (ICCV 2017 · similarity 0.91)
- Camera-Aware Proxies for Unsupervised Person Re-Identification (AAAI 2021 · similarity 0.90)
- Embedding Deep Metric for Person Re-identification: A Study Against Large Variations (ECCV 2016 · similarity 0.90)
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.
