Research map / Video / Action recognition
Recognizing Actions in 3D Using Action-Snippets and Activated Simplices
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| activated simplices instead of nearest-neighbor on action-snippets (ablation) | action recognition · MSR-Action3D | accuracy: 85.16% -> 91.40% | Helped |
| activated simplices instead of k-means (ablation) | action recognition · MSR-Action3D | accuracy: 79.30% -> 91.40% | Helped |
| activated simplices instead of sparse coding (ablation) | action recognition · MSR-Action3D | accuracy: 86.30% -> 91.40% | Helped |
| add non-negative sum-to-one constraint to sparse coding (ablation) | sparse coding · MSR-Action3D | accuracy: 86.30% -> 86.94% | No clear effect |
| stack more poses into each action-snippet (up to 10) (ablation) | activated simplices · MSR-Action3D | accuracy | Helped |
| increase number of bases per class beyond 50 (ablation) | activated simplices · MSR-Action3D | accuracy | No clear effect |
| increase number of bases beyond 20 (ablation) | sparse coding · MSR-Action3D | accuracy | Hurt |
| activated simplices instead of sparse coding and k-means for partially observed poses (depth removed) (ablation) | pose reconstruction · H3.6M | reconstruction error | Helped |
Closest papers by meaning
- Learning Discriminative Activated Simplices for Action Recognition (AAAI 2017 · similarity 0.91)
- Mining 3D Key-Pose-Motifs for Action Recognition (CVPR 2016 · similarity 0.89)
- Rolling Rotations for Recognizing Human Actions From 3D Skeletal Data (CVPR 2016 · similarity 0.87)
- Recovering Complete Actions for Cross-dataset Skeleton Action Recognition (NeurIPS 2024 · similarity 0.84)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.84)
- Skeleton-based Action Recognition with Non-linear Dependency Modeling and Hilbert-Schmidt Independence Criterion (AAAI 2025 · similarity 0.84)
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.