Research map / Video / Action recognition
Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations
Kezhen Chen, Kenneth Forbus
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 |
|---|---|---|---|
| select extra qualitative direction features with dynamic feature selection instead of manually chosen ones (ablation) | analogical generalization pipeline · Florence 3D Actions, cross-subject | accuracy (%): 63.6 -> 74.2 | Improved |
| restrict UTD-MHAD to 19 actions by removing 6 similar and 2 noisy actions (ablation) | analogical generalization pipeline · UTD-MHAD cross-subject | accuracy (%): 65.82 -> 80.3 | Improved |
| use analogical generalization over qualitative relations instead of CNN on joint trajectory maps | skeleton action recognition · UTD-MHAD cross-subject | accuracy (%): 85.81 -> 65.82 | Hurt |
| use analogical generalization instead of Lie-group representation with SVM | skeleton action recognition · Florence 3D Actions LOOCV | accuracy (%): 90.88 -> 86.9 | Hurt |
| use analogical generalization instead of sparse representation in dissimilarity space | skeleton action recognition · UTKinect-Action3D LOOCV | accuracy (%): 90.95 -> 88.50 | Hurt |
Closest papers by meaning
- InfoGCN: Representation Learning for Human Skeleton-Based Action Recognition (CVPR 2022 · similarity 0.90)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.88)
- Recovering Complete Actions for Cross-dataset Skeleton Action Recognition (NeurIPS 2024 · similarity 0.88)
- Skeleton-based Action Recognition with Non-linear Dependency Modeling and Hilbert-Schmidt Independence Criterion (AAAI 2025 · similarity 0.88)
- Stitch, Contrast, and Segment: Learning a Human Action Segmentation Model Using Trimmed Skeleton Videos (AAAI 2025 · similarity 0.87)
- Heterogeneous Skeleton-Based Action Representation Learning (CVPR 2025 · similarity 0.87)
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.
