Research map / Label-efficient and robust learning / Recognition and long-tail
SOSNet: Second Order Similarity Regularization for Local Descriptor Learning
Yurun Tian, Xin Yu, Bin Fan, Fuchao Wu, Huub Heijnen, Vassileios Balntas
Check 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 SOSR to the descriptor training objective (ablation) | SOSNet training · UBC Phototour, Liberty-trained models evaluated on Notredame and Yosemite | FPR@95: 19.49% | Improved |
| Replace Hinge Triplet loss with Quadratic Hinge Triplet loss (ablation) | Descriptor training with SGD or Adam · UBC Phototour, Liberty-trained models evaluated on Notredame and Yosemite | FPR@95: 5.46% | Improved |
| Use Adam instead of SGD (ablation) | Descriptor training · UBC Phototour, Liberty-trained models evaluated on Notredame and Yosemite | FPR@95: 11.63% | Improved |
| Use SOSR instead of Global Orthogonal Regularization (ablation) | Descriptor training · UBC Phototour, Liberty-trained models evaluated on Notredame and Yosemite | FPR@95 | Improved |
| Set batch pair count N to 512 and nearest-neighbor count K to 8 (ablation) | SOSNet · UBC Phototour, Liberty-trained models evaluated on Notredame and Yosemite | mean FPR@95 | Improved |
| Exclude far-away negatives and calculate SOSR over nearest neighbors (ablation) | SOSNet · Descriptor training | matching performance | Improved |
Later papers that compared with this method
2 later accepted papers report SOSNet as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Descriptor Distillation: A Teacher-Student-Regularized Framework for Learning Local Descriptors (IJCV 2024 · better on 1 of 1 shared result)
- Online Invariance Selection for Local Feature Descriptors (ECCV 2020 · better on 4 of 6 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
- Gromov-Wasserstein Learning for Graph Matching and Node Embedding (ICML 2019 · similarity 0.86)
- Efficient Graph Similarity Computation with Alignment Regularization (NeurIPS 2022 · similarity 0.85)
- Descriptor Distillation: A Teacher-Student-Regularized Framework for Learning Local Descriptors (IJCV 2024 · similarity 0.85)
- Local Descriptors Optimized for Average Precision (CVPR 2018 · similarity 0.84)
- Cluster-Aware Similarity Diffusion for Instance Retrieval (ICML 2024 · similarity 0.83)
- THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplings (AAAI 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.
