Research map / Label-efficient and robust learning / Recognition and long-tail
Coded Residual Transform for Generalizable Deep Metric Learning
SHICHAO KAN, Yixiong Liang, Min Li, Yigang Cen, Jianxin Wang, Zhihai He
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 |
|---|---|---|---|
| apply CRT (ablation) | MS metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 73.97 -> 75.95 | Improved |
| add CRT-Consistency (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 72.25 -> 75.95 | Improved |
| add multi-perspective CRT feature transformation (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 74.34 -> 75.95 | Improved |
| use different embedding networks for each prototype's CRT feature instead of the same embedding network (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1 | Improved |
| vary first-branch prototype count from 1 to 4, 16, or 64 (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 75.76; 75.57; 75.44; 75.96 | Mixed |
| vary second-branch prototype count from 1 to 4, 16, 64, or 100 (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 74.26; 74.54; 74.63; 75.95; 75.15 | Mixed |
| share weights between the two embedding branches (ablation) | CRT metric learning with MiT-B1 · CUB-200-2011 | Recall@1: 75.96 vs 75.95 | No clear effect |
| replace Hyp-DeiT with CRT | deep metric learning · CUB-200-2011 | Recall@1: 74.7 -> 78.98 | Improved |
| replace Group Loss++ with CRT | deep metric learning · Cars-196 | Recall@1: 90.4 -> 91.16 | Improved |
| replace XBM+RTT with CRT | deep metric learning · SOP | Recall@1: 84.5 -> 83.41 | Hurt |
| replace IBC with CRT | deep metric learning · In-Shop | Recall@1: 92.8 -> 94.48 | Improved |
Closest papers by meaning
- Sharing Matters for Generalization in Deep Metric Learning (TPAMI 2020 · similarity 0.86)
- Multi-Similarity Loss With General Pair Weighting for Deep Metric Learning (CVPR 2019 · similarity 0.85)
- Spherical Feature Transform for Deep Metric Learning (ECCV 2020 · similarity 0.85)
- Signal-To-Noise Ratio: A Robust Distance Metric for Deep Metric Learning (CVPR 2019 · similarity 0.85)
- Deep Disentangled Metric Learning (AAAI 2025 · similarity 0.85)
- Improving Deep Metric Learning by Divide and Conquer (TPAMI 2021 · 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.
