Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Relational Metric Learning
Wenzhao Zheng, Borui Zhang, Jiwen Lu, Jie Zhou
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 |
|---|---|---|---|
| add DRML to ProxyAnchor | ProxyAnchor · CUB-200-2011 | Recall@1: 67.3 -> 68.7 | Improved |
| add DRML to ProxyAnchor | ProxyAnchor · Cars196 | Recall@1: 84.4 -> 86.9 | Improved |
| add DRML to margin loss with distance-weighted sampling | Margin-DW · Stanford Online Products | Recall@1: 76.8 -> 79.9 | Improved |
| add DRML to triplet loss with semi-hard sampling | Triplet-SH · Stanford Online Products | Recall@1: 76.1 -> 75.7 | Hurt |
| add DRML to triplet loss with semi-hard sampling | Triplet-SH · CUB-200-2011 | Recall@1: 58.8 -> 61.7 | Improved |
| add DRML to ProxyAnchor | ProxyAnchor · ImageNet | R/P@1: 66.4 -> 68.0 | Improved |
| add relational module to adaptive ensemble (ablation) | AEL-PA · CUB-200-2011 | Recall@1: 60.5 -> 68.7 | Improved |
| add relational module to DiVA ensemble (ablation) | DiVA · CUB-200-2011 | Recall@1: 66.4 -> 63.8 | Hurt |
| add ensemble loss J_ensem (ablation) | DRML · CUB-200-2011 | Recall@1: 45.6 -> 68.8 | Improved |
| add embedding loss J_emb (ablation) | DRML · CUB-200-2011 | Recall@1: 60.4 -> 68.8 | Improved |
| add reconstruction loss J_recon (ablation) | DRML · CUB-200-2011 | Recall@1: 66.6 -> 68.8 | Improved |
| use learned relations instead of random fixed relations (ablation) | DRML-PA · CUB-200-2011 | Recall@1: 62.6 -> 68.7 | Improved |
Closest papers by meaning
- Deep Metric Learning with Adaptively Composite Dynamic Constraints (TPAMI 2023 · similarity 0.90)
- Improving Deep Metric Learning by Divide and Conquer (TPAMI 2021 · similarity 0.90)
- Deep Metric Learning via Adaptive Learnable Assessment (CVPR 2020 · similarity 0.88)
- Divide and Conquer the Embedding Space for Metric Learning (CVPR 2019 · similarity 0.88)
- Deep Disentangled Metric Learning (AAAI 2025 · similarity 0.87)
- Towards Improved Proxy-Based Deep Metric Learning via Data-Augmented Domain Adaptation (AAAI 2024 · 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.
