Research map / Video / Action recognition
Spatio-Temporal Difference Descriptor for Skeleton-Based Action Recognition
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 |
|---|---|---|---|
| Add the spatio-temporal difference descriptor (ablation) | LSTM · G3D-Gaming | accuracy: 80.97% -> 87.56% | Helped |
| Add directional convolution (ablation) | StddNet · G3D-Gaming | accuracy: 87.56% -> 88.97%; 89.23%; 90.06% | Helped |
| Increase directional partitions from b = 8 to b = 16 (ablation) | Di-StddNet · G3D-Gaming | accuracy: 88.97% -> 90.06% | Helped |
| Add non-linear mapping blocks, increasing from 0 to 2 (ablation) | Di-StddNet (b = 16) · G3D-Gaming | accuracy: 90.06% -> 92.03% | Helped |
| Increase mapping blocks from 2 to 3 (ablation) | Di-StddNet (b = 16) · G3D-Gaming | accuracy: 92.03% -> 89.58% | Hurt |
| Add mapping blocks from 0 to 2 (ablation) | Di-StddNet · HDM05 | accuracy: 74.60%±2.12 -> 77.83%±3.07; 76.43%±2.25 -> 80.27%±2.35; 77.58%±2.98 -> 82.32%±2.27 | Helped |
| Add mapping blocks from 0 to 2 (ablation) | Di-StddNet · Northwestern-UCLA | accuracy: 87.3% -> 90.6%; 88.4% -> 93.8%; 89.8% -> 92.4% | Helped |
| Use b = 12 instead of b = 16 with 2 mapping blocks (ablation) | Di-StddNet · Northwestern-UCLA | accuracy: 92.4% -> 93.8% | Helped |
Closest papers by meaning
- Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition (AAAI 2018 · similarity 0.91)
- Graph CNNs with Motif and Variable Temporal Block for Skeleton-Based Action Recognition (AAAI 2019 · similarity 0.90)
- Towards To-a-T Spatio-Temporal Focus for Skeleton-Based Action Recognition (AAAI 2022 · similarity 0.90)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.89)
- Skeleton-based Action Recognition with Non-linear Dependency Modeling and Hilbert-Schmidt Independence Criterion (AAAI 2025 · similarity 0.89)
- Heterogeneous Skeleton-Based Action Representation Learning (CVPR 2025 · similarity 0.88)
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.