Research map / Label-efficient and robust learning / Recognition and long-tail
DARI: Distance Metric and Representation Integration for Person Verification
Guangrun Wang, Liang Lin, Shengyong Ding, Ya Li, Qing Wang
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 |
|---|---|---|---|
| jointly optimize the distance metric and CNN-based representation (ablation) | DARI · CUHK03 | rank-1 recognition rate: 45.4% -> 55.4% | Improved |
| jointly optimize the distance metric and CNN-based representation (ablation) | DARI · CUHK01 | rank-1 recognition rate: 57.7% -> 65.95% | Improved |
| incorporate the Mahalanobis distance matrix (ablation) | DARI · iLIDS | top-1 performance: 36.9% -> 42.8% | Improved |
| add center cropping with random perturbation for data augmentation (ablation) | DARI · iLIDS | top-1 performance: drop by about 30 percent relatively without augmentation | Improved |
| construct 4800 rather than 60 triplets per iteration for the same 60 persons (ablation) | DARI · iLIDS | performance: about 70 percent of the second strategy | Improved |
Closest papers by meaning
- Embedding Deep Metric for Person Re-identification: A Study Against Large Variations (ECCV 2016 · similarity 0.87)
- Cross-View Asymmetric Metric Learning for Unsupervised Person Re-Identification (ICCV 2017 · similarity 0.85)
- Unsupervised Person Re-Identification by Deep Asymmetric Metric Embedding (TPAMI 2018 · similarity 0.85)
- Person Re-Identification by Camera Correlation Aware Feature Augmentation (TPAMI 2017 · similarity 0.85)
- Hard-Aware Point-to-Set Deep Metric for Person Re-identification (ECCV 2018 · similarity 0.85)
- CA-Jaccard: Camera-aware Jaccard Distance for Person Re-identification (CVPR 2024 · similarity 0.83)
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.
