Research map / Graph learning / Knowledge and temporal graphs
Structure-Aware Human-Action Generation
Ping Yu, Yang Zhao, Chunyuan Li, Junsong Yuan, Changyou Chen
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 5 GCN layers instead of 5 CNN layers (ablation) | SA-GCN · Human3.6m (ablation) | MMD_avg: 0.240 -> 0.152 | Improved |
| add 5 GCN layers instead of 5 CNN layers (ablation) | SA-GCN · Human3.6m (ablation) | MMD_seq: 0.222 -> 0.142 | Improved |
| add top-k attention connections to past frames (ablation) | SA-GCN · Human3.6m (ablation) | MMD_avg: 0.915 -> 0.152 | Improved |
| add top-k attention connections to past frames (ablation) | SA-GCN · Human3.6m (ablation) | MMD_seq: 0.922 -> 0.142 | Improved |
| use SA-GC layers instead of ST-GC layers (ablation) | SA-GCN · Human3.6m (ablation) | MMD_avg: 0.580 -> 0.152 | Improved |
| use SA-GC layers instead of ST-GC layers (ablation) | SA-GCN · Human3.6m (ablation) | MMD_seq: 0.595 -> 0.142 | Improved |
| SA-GCN trained without pretraining vs pretrained c-SkeletonGAN | SA-GCN · Human3.6m | MMD_avg: 0.195 -> 0.146 | Improved |
| SA-GCN vs c-SkeletonGAN on recognition accuracy | SA-GCN · Human3.6m | recognition accuracy: 0.488 -> 0.593 | Improved |
| SA-GCN vs c-SkeletonGAN on NTU cross-view | SA-GCN · NTU RGB+D (cross-view) | MMD_avg: 0.371 -> 0.316 | Improved |
| SA-GCN vs c-SkeletonGAN on NTU cross-subject | SA-GCN · NTU RGB+D (cross-subject) | MMD_avg: 0.338 -> 0.285 | Improved |
| SA-GCN vs c-SkeletonGAN in human evaluation | SA-GCN · Human3.6m human evaluation (AMT) | evaluation score: 3.157 -> 3.925 | Improved |
Closest papers by meaning
- Actional-Structural Graph Convolutional Networks for Skeleton-Based Action Recognition (CVPR 2019 · similarity 0.92)
- Learning Graph Convolutional Network for Skeleton-Based Human Action Recognition by Neural Searching (AAAI 2020 · similarity 0.91)
- 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 0.89)
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.
