Research map / 3D vision / Human pose estimation
Action2video: Generating Videos of Human 3D Actions
Chuan Guo, Xinxin Zuo, Sen Wang, Xinshuang Liu, Shihao Zou, Minglun Gong, Li Cheng
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 Lie algebra pose representation (ablation) | Action2Motion (plain) · HumanAct12 (coarse-grained) | FID: 3.299 -> 2.458 | Improved |
| add Lie algebra pose representation (ablation) | Action2Motion (plain) · HumanAct12 (fine-grained) | FID: 1.329 -> 1.000 | Improved |
| add Lie algebra pose representation (ablation) | Action2Motion (plain) · HumanAct12 (coarse-grained) | accuracy: 0.656 -> 0.923 | Improved |
| add Lie algebra pose representation (ablation) | Action2Motion (plain) · CMU MoCap | FID: 2.994 -> 2.885 | No clear effect |
| add Lie algebra pose representation (ablation) | Action2Motion (plain) · CMU MoCap | accuracy: 0.378 -> 0.686 | Improved |
| add Lie algebra pose representation (ablation) | Action2Motion (plain) · NTU-RGBD | FID: 0.540 -> 0.330 | Improved |
| add global and local movement integration (MLP) (ablation) | Action2Motion (w/ Lie) · HumanAct12 (coarse-grained) | FID: 2.458 -> 2.157 | Improved |
| add global and local movement integration (MLP) (ablation) | Action2Motion (w/ Lie) · HumanAct12 (fine-grained) | FID: 1.000 -> 0.739 | Improved |
| add global and local movement integration (MLP) (ablation) | Action2Motion (w/ Lie) · CMU MoCap | FID: 2.885 -> 2.448 | Improved |
| add global and local movement integration (MLP) (ablation) | Action2Motion (w/ Lie) · HumanAct12 (coarse-grained) | accuracy: 0.923 -> 0.835 | Hurt |
| implement GLMI with GRU instead of MLP (ablation) | Action2Motion (GLMI-M) · HumanAct12 (coarse-grained) | FID: 2.157 -> 2.349 | Hurt |
| implement GLMI with GRU instead of MLP (ablation) | Action2Motion (GLMI-M) · HumanAct12 (fine-grained) | FID: 0.739 -> 0.957 | Hurt |
Closest papers by meaning
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.90)
- Motion-2-to-3: Leveraging 2D Motion Data for 3D Motion Generations (ICCV 2025 · similarity 0.90)
- Learning 3D Human Dynamics From Video (CVPR 2019 · similarity 0.90)
- Move-in-2D: 2D-Conditioned Human Motion Generation (CVPR 2025 · similarity 0.89)
- VideoMDM: Towards 3D Human Motion Generation From 2D Supervision (NeurIPS 2026 · similarity 0.89)
- HumanDreamer: Generating Controllable Human-Motion Videos via Decoupled Generation (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.
