Research map / Label-efficient and robust learning / Self-supervised and contrastive
Deep Metric Learning with Self-Supervised Ranking
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 |
|---|---|---|---|
| Add the self-supervised ranking framework to Margin loss | ResNet50 metric learning · CUB-200-2011 | R@1: 63.4 -> 66.5 | Helped |
| Add the self-supervised ranking framework to Margin loss | ResNet50 metric learning · Cars-196 | R@1: 79.5 -> 84.5 | Helped |
| Add the self-supervised ranking framework to Margin loss | ResNet50 metric learning · Stanford Online Products | R@1: 68.0 -> 70.1 | Helped |
| Add the self-supervised ranking framework to MS loss | ResNet50 metric learning · Cars-196 | R@1: 81.6 -> 85.5 | Helped |
| Add the self-supervised ranking framework to MS loss | ResNet50 metric learning · Stanford Online Products | R@1: 69.8 -> 70.5 | Helped |
| Add the ranking-sort loss Lsort (ablation) | Self-supervised ranking framework with Margin loss · CUB-200-2011 | R@1: 61.9 -> 66.5 | Helped |
| Add the ranking-sort loss Lsort (ablation) | Self-supervised ranking framework with Margin loss · Cars-196 | R@1: 78.3 -> 84.5 | Helped |
| Add the positive-pair boundary loss Lpos (ablation) | Self-supervised ranking framework with Margin loss · CUB-200-2011 | R@1: 60.7 -> 66.5 | Helped |
| Add the positive-pair boundary loss Lpos (ablation) | Self-supervised ranking framework with Margin loss · Cars-196 | R@1: 76.5 -> 84.5 | Helped |
| Add the ranking-task MLP head gθ (ablation) | Self-supervised ranking framework with Margin loss · CUB-200-2011 | R@1: 62.5 -> 66.5 | Helped |
| Add the ranking-task MLP head gθ (ablation) | Self-supervised ranking framework with Margin loss · Cars-196 | R@1: 79.4 -> 84.5 | Helped |
| Use both Lsort and Lpos instead of Lsort alone (ablation) | Self-supervised ranking framework with Margin loss · CUB-200-2011 | R@1: 62.5 -> 64.3 | Helped |
Closest papers by meaning
- Ranked List Loss for Deep Metric Learning (CVPR 2019 · similarity 0.90)
- Divide and Conquer the Embedding Space for Metric Learning (CVPR 2019 · similarity 0.89)
- SLADE: A Self-Training Framework for Distance Metric Learning (CVPR 2021 · similarity 0.88)
- Self-Taught Metric Learning Without Labels (CVPR 2022 · similarity 0.88)
- Deep Metric Learning via Lifted Structured Feature Embedding (CVPR 2016 · similarity 0.87)
- Neighborhood-Adaptive Structure Augmented Metric Learning (AAAI 2022 · similarity 0.87)
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.