Research map / Graph learning / Knowledge and temporal graphs
Neural Graph Matching Networks for Fewshot 3D Action Recognition
Michelle Guo, Edward Chou, De-An Huang, Shuran Song, Serena Yeung, Li Fei-Fei
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 learned edges (vs no edges) (ablation) | NGM · CAD-120 (1-shot) | accuracy (%): 66.1 -> 78.5 | Improved |
| add learned edges (vs no edges) (ablation) | NGM · CAD-120 (5-shot) | accuracy (%): 85.0 -> 91.1 | Improved |
| add learned edges (vs no edges) (ablation) | NGM · PiGraphs (1-shot) | accuracy (%): 75.9 -> 80.2 | Improved |
| add learned edges (vs no edges) (ablation) | NGM · PiGraphs (5-shot) | accuracy (%): 71.1 -> 88.3 | Improved |
| use learned edges instead of Human-Object heuristic edges (ablation) | NGM · CAD-120 (1-shot) | accuracy (%): 74.9 -> 78.5 | Improved |
| use learned edges instead of Human-Object heuristic edges (ablation) | NGM · PiGraphs (5-shot) | accuracy (%): 71.3 -> 88.3 | Improved |
| use learned edges instead of Proximity heuristic edges (ablation) | NGM · CAD-120 (5-shot) | accuracy (%): 88.1 -> 91.1 | Improved |
| Human-Object heuristic edges instead of no edges (ablation) | NGM · PiGraphs (5-shot) | accuracy (%): 82.5 -> 71.3 | Hurt |
| Human-Object heuristic edges instead of no edges (ablation) | NGM · CAD-120 (1-shot) | accuracy (%): 66.1 -> 74.9 | Improved |
| add adjacency matching to feature matching in the graph tensor (ablation) | NGM · CAD-120 (1-shot) | accuracy (%): 61.1 -> 78.5 | Improved |
| add feature matching to adjacency-only matching (ablation) | NGM · PiGraphs (5-shot) | accuracy (%): 74.1 -> 88.3 | Improved |
| add adjacency matching to feature matching (ablation) | NGM · PiGraphs (1-shot) | accuracy (%): 77.8 -> 80.2 | Improved |
Closest papers by meaning
- Boosting Few-shot Action Recognition with Graph-guided Hybrid Matching (ICCV 2023 · similarity 0.90)
- Few-Shot Learning with Graph Neural Networks (ICLR 2018 · similarity 0.87)
- A Structured Model for Action Detection (CVPR 2019 · similarity 0.86)
- I Know the Relationships: Zero-Shot Action Recognition via Two-Stream Graph Convolutional Networks and Knowledge Graphs (AAAI 2019 · similarity 0.86)
- Symbiotic Graph Neural Networks for 3D Skeleton-Based Human Action Recognition and Motion Prediction (TPAMI 2021 · similarity 0.85)
- Task-Specific Distance Correlation Matching for Few-Shot Action Recognition (AAAI 2026 · 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.
