Research map / Label-efficient and robust learning / Domain adaptation and generalization
OVL: One-View Learning for Human Retrieval
Wenjing Li, Zhongcheng Wu
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 |
|---|---|---|---|
| remove triplet loss (ablation) | OVL framework · Market-1501 avg rank-1 | Rank-1: 59.5 -> 67.8 | Improved |
| add adversarial multi-view learning (AMVL) (ablation) | OVL framework · Market-1501 avg rank-1 | Rank-1: 61.7 -> 67.8 | Improved |
| add adversarial unknown rejection learning (AURL) (ablation) | OVL framework · Market-1501 avg rank-1 | Rank-1: 62.0 -> 67.8 | Improved |
| add AURL only to baseline (no AMVL) (ablation) | Baseline · Market-1501 avg rank-1 | Rank-1: 41.9 -> 61.7 | Improved |
| AMVL instead of DANN (align to global target view) | Baseline · Market-1501 avg rank-1 | Rank-1: 52.4 -> 62.0 | Improved |
| AMVL instead of ADDA with C-1 domain classifiers | Baseline · Market-1501 avg rank-1 | Rank-1: 55.6 -> 62.0 | Improved |
| lambda_vi from 0.0 to 0.2 (ablation) | OVL framework · Market-1501 (3rd view) | Rank-1: 72.2 -> 78.1 | Improved |
| lambda_vi large (1.0) (ablation) | OVL framework · Market-1501 (3rd view) | Rank-1: 78.1 -> 73.5 | Hurt |
| lambda_un from 0.0 to 0.1 (ablation) | OVL framework · Market-1501 (3rd view) | Rank-1: 70.3 -> 78.1 | Improved |
| lambda_un increased to 0.5 (ablation) | OVL framework · Market-1501 (3rd view) | Rank-1: 78.1 -> 73.0 | Hurt |
| full method vs source-view-only baseline | Baseline · DukeMTMC-reID avg rank-1 | Rank-1: 27.6 -> 52.1 | Improved |
| Ours vs domain adaptation method HHL | OVL framework · Market-1501 | Rank-1: 62.2 -> 78.1 | Improved |
Closest papers by meaning
- Deep Neural Networks with Inexact Matching for Person Re-Identification (NeurIPS 2016 · similarity 0.86)
- Joint Generative and Contrastive Learning for Unsupervised Person Re-Identification (CVPR 2021 · similarity 0.86)
- Unsupervised Person Re-Identification by Soft Multilabel Learning (CVPR 2019 · similarity 0.86)
- Image-Image Domain Adaptation With Preserved Self-Similarity and Domain-Dissimilarity for Person Re-Identification (CVPR 2018 · similarity 0.86)
- Person Re-Identification by Camera Correlation Aware Feature Augmentation (TPAMI 2017 · similarity 0.85)
- Adversarially Occluded Samples for Person Re-Identification (CVPR 2018 · similarity 0.85)
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.
