Research map / Graph learning / Knowledge and temporal graphs
Decoupling GCN with DropGraph Module for Skeleton-Based Action Recognition
Ke Cheng, Yifan Zhang, Congqi Cao, Lei Shi, Jian Cheng, Hanqing Lu
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 decoupling graph convolution (ablation) | ST-GCN · NW-UCLA | accuracy (%): 89.8 -> 91.6 | Improved |
| add decoupling graph convolution (ablation) | ST-GCN · NTU-RGBD-120 X-sub | accuracy (%): 79.7 -> 81.3 | Improved |
| add decoupling graph convolution (ablation) | ST-GCN · NTU-RGBD-120 X-setup | accuracy (%): 81.3 -> 82.7 | Improved |
| add attention-guided DropGraph on top of decoupling (ablation) | ST-GCN+DC · NW-UCLA | accuracy (%): 91.6 -> 93.8 | Improved |
| add attention-guided DropGraph on top of decoupling (ablation) | ST-GCN+DC · NTU-RGBD-120 X-sub | accuracy (%): 81.3 -> 82.4 | Improved |
| add attention-guided DropGraph on top of decoupling (ablation) | ST-GCN+DC · NTU-RGBD-120 X-setup | accuracy (%): 82.7 -> 84.3 | Improved |
| add dropout (ablation) | DC-GCN · NTU-RGBD X-sub | top-1 accuracy (%): 87.1 -> 87.2 | No clear effect |
| add label smoothing (ablation) | DC-GCN · NTU-RGBD X-view | top-1 accuracy (%): 94.3 -> 94.4 | No clear effect |
| add DropEdge (ablation) | DC-GCN · NTU-RGBD X-sub | top-1 accuracy (%): 87.1 -> 87.6 | Improved |
| add DropGraph (ablation) | DC-GCN · NTU-RGBD X-sub | top-1 accuracy (%): 87.1 -> 88.0 | Improved |
| add DropGraph (ablation) | DC-GCN · NTU-RGBD X-view | top-1 accuracy (%): 94.3 -> 95.0 | Improved |
| add attention guidance to DropGraph (ablation) | DropGraph · NTU-RGBD X-sub | top-1 accuracy (%): 88.0 -> 88.2 | Improved |
Closest papers by meaning
- Skeleton-Based Action Recognition With Shift Graph Convolutional Network (CVPR 2020 · similarity 0.92)
- Learning Graph Convolutional Network for Skeleton-Based Human Action Recognition by Neural Searching (AAAI 2020 · similarity 0.92)
- Are Spatial-Temporal Graph Convolution Networks for Human Action Recognition Over-Parameterized? (CVPR 2025 · similarity 0.90)
- BlockGCN: Redefine Topology Awareness for Skeleton-Based Action Recognition (CVPR 2024 · similarity 0.90)
- Dynamic Semantic-Based Spatial Graph Convolution Network for Skeleton-Based Human Action Recognition (AAAI 2024 · similarity 0.90)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (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.
