Research map / Label-efficient and robust learning / Recognition and long-tail
Learning With Memory-Based Virtual Classes for Deep Metric Learning
Byungsoo Ko, Geonmo Gu, Han-Gyu Kim
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 MemVir virtual classes | Softmax · CARS196 (conventional) | Recall@1: 81.5 -> 86.5 (+5.0) | Improved |
| add MemVir virtual classes | Norm-softmax · CUB200 (conventional) | Recall@1: 64.9 -> 67.3 (+2.4) | Improved |
| add MemVir virtual classes | Proxy-anchor · SOP (conventional) | Recall@1: 78.9 -> 79.7 (+0.8) | Improved |
| add MemVir virtual classes | Proxy-NCA · CUB200 (conventional) | Recall@1: 64.3 -> 68.3 (+4.0) | Improved |
| add MemVir virtual classes | Proxy-anchor · CARS196 (MLRC) | P@1: 86.20 -> 86.40 | Improved |
| add MemVir virtual classes | ArcFace · SOP (MLRC) | P@1: 76.20 -> 76.05 | No clear effect |
| add MemVir virtual classes | Norm-softmax · CARS196 batch size 1024 | Recall@1: 78.5 -> 84.6 (+6.1) | Improved |
| add MemVir virtual classes | Norm-softmax · CARS196 10% classes | Recall@1: 56.4 -> 58.5 (+2.1) | Improved |
| use Virtual softmax instead of softmax | SoftMax · CARS196 (ResNet50) | Recall@1: 78.3 -> 75.1 | Hurt |
| add XBM memory to ArcFace | ArcFace · CARS196 (ResNet50) | Recall@1: 78.8 -> 78.9 | No clear effect |
| use MemVir instead of BroadFace on ArcFace | ArcFace · CARS196 (ResNet50) | Recall@1: 79.5 -> 80.7 | Improved |
| add MemVir virtual classes | CurricularFace · SOP (ResNet50) | Recall@1: 79.8 -> 81.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ArcFace — ArcFace: Additive Angular Margin Loss for Deep Face Recognition (CVPR 2019) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Revisiting Training Strategies and Generalization Performance in Deep Metric Learning (ICML 2020 · similarity 0.88)
- Improving Deep Metric Learning by Divide and Conquer (TPAMI 2021 · similarity 0.87)
- Towards Improved Proxy-Based Deep Metric Learning via Data-Augmented Domain Adaptation (AAAI 2024 · similarity 0.85)
- Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning (ICLR 2024 · similarity 0.85)
- Deep Disentangled Metric Learning (AAAI 2025 · similarity 0.84)
- Potential Field Based Deep Metric Learning (CVPR 2025 · 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.
