Research map / Label-efficient and robust learning / Recognition and long-tail
Maximum Margin Metric Learning Over Discriminative Nullspace for Person Re-identification
T M Feroz Ali, Subhasis Chaudhuri
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 NK3ML instead of NFST | WHOS features · PRID450S | Rank-1 accuracy (%): 42.58 -> 50.67 | Improved |
| use NK3ML instead of KNFST | LOMO features · GRID | Rank-1 accuracy (%): 14.88 -> 18.24 | Improved |
| use NK3ML instead of XQDA | GOG features · GRID | Rank-1 accuracy (%): 24.80 -> 26.96 | Improved |
| use NK3ML instead of XQDA | WHOS features · PRID450S | Rank-1 accuracy (%): 43.38 -> 50.67 | Improved |
| use NK3ML instead of NFST | GOG features · PRID450S | Rank-1 accuracy (%): 67.60 -> 68.04 | No clear effect |
| use NK3ML instead of XQDA | GOG features · PRID450S | Rank-1 accuracy (%): 68.00 -> 68.04 | No clear effect |
| use NK3ML instead of KNFST | WHOS features · GRID | Rank-1 accuracy (%): 21.12 -> 21.20 | No clear effect |
| use NK3ML instead of GOG+XQDA | CUHK01 single-shot · CUHK01 | Rank-1 accuracy (%): 57.89 -> 67.09 | Improved |
| use NK3ML instead of GOG+XQDA | CUHK01 multi-shot · CUHK01 | Rank-1 accuracy (%): 67.28 -> 76.77 | Improved |
| use NK3ML instead of GOG+XQDA | PRID450S · PRID450S | Rank-1 accuracy (%): 68.00 -> 73.42 | Improved |
| use NK3ML instead of the previous best (DCIA, with post-processing) | VIPeR · VIPeR | Rank-1 accuracy (%): 63.9 -> 99.8 | Improved |
| use NK3ML instead of NFST (training cost) | VIPeR · VIPeR | Training time (s): 1.47 -> 1.64 | No clear effect |
Closest papers by meaning
- Learning a Discriminative Null Space for Person Re-Identification (CVPR 2016 · similarity 0.91)
- Efficient Online Local Metric Adaptation via Negative Samples for Person Re-Identification (ICCV 2017 · similarity 0.88)
- Learning Incremental Triplet Margin for Person Re-Identification (AAAI 2019 · similarity 0.87)
- Sample-Specific SVM Learning for Person Re-Identification (CVPR 2016 · similarity 0.86)
- Graph Sampling Based Deep Metric Learning for Generalizable Person Re-Identification (CVPR 2022 · similarity 0.86)
- Embedding Deep Metric for Person Re-identification: A Study Against Large Variations (ECCV 2016 · similarity 0.86)
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.
