Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Structured Set Matching Networks for One-Shot Part Labeling
Jonghyun Choi, Jayant Krishnamurthy, Aniruddha Kembhavi, Ali Farhadi
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 global consistency and global normalization to set matching (ablation) | SSMN · DiPART | accuracy: 44.7% -> 58.1% | Improved |
| add global consistency and global normalization to set matching (ablation) | SSMN · PPM | accuracy: 40.6% -> 46.6% | Improved |
| remove the part appearance factor (ablation) | SSMN · Cross-DiPART-PPM | accuracy: 30.8% -> 33.1% | Improved |
| remove the part appearance factor (ablation) | SSMN · DiPART validation | accuracy: within 0.1% | No clear effect |
| use distance transform images instead of original diagrams (ablation) | Appearance Matching Network · diagram appearance matching | accuracy: 33.5% -> 38.4% | Improved |
| use 3x3 filters instead of 15x15 filters (ablation) | CNN encoder for distance transform images · diagram appearance matching | accuracy | Mixed |
| allow unmatched parts with different-sized source and target sets | SSMN · DiPART, 9 source and 8 target parts | test accuracy: 31.5% -> 38.2% | Improved |
| use globally normalized rather than locally normalized training (ablation) | SSMN · DiPART and PPM | accuracy | Improved |
| use a larger inference beam (ablation) | SSMN · DiPART and PPM | accuracy | No clear effect |
Closest papers by meaning
- Graph Structured Network for Image-Text Matching (CVPR 2020 · similarity 0.88)
- Learning to Detect Semantic Boundaries with Image-Level Class Labels (IJCV 2022 · similarity 0.86)
- GlueStick: Robust Image Matching by Sticking Points and Lines Together (ICCV 2023 · similarity 0.86)
- Graph-Structured Referring Expression Reasoning in the Wild (CVPR 2020 · similarity 0.86)
- Mapping Images to Scene Graphs with Permutation-Invariant Structured Prediction (NeurIPS 2018 · similarity 0.86)
- Visual Boundary Knowledge Translation for Foreground Segmentation (AAAI 2021 · 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.
