Research map / 3D vision / Human pose estimation
HDG-ODE: A Hierarchical Continuous-Time Model for Human Pose Forecasting
Yucheng Xing, Xin Wang
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 |
|---|---|---|---|
| use continuous-time ODE propagation instead of discrete-time recurrence (ablation) | graph pose forecasting · MuPoTS-3D | MPJPE: 0.4375 -> 0.4066 (SemGCGRU vs Graph-ODE) | Improved |
| use relational top-down pipeline instead of isolated per-person pipeline (ablation) | Graph-GRU · MuPoTS-3D | MPJPE: 0.5738 -> 0.5573 | Improved |
| use relational pipeline instead of isolated (ablation) | STGCN · MuPoTS-3D | PCKh@1.0: 77.72% -> 77.76% | No clear effect |
| use time-varying dynamic graph convolution instead of static graph convolution (ablation) | Graph-GRU · MuPoTS-3D | MPJPE: 0.5573 -> 0.3844 | Improved |
| use time-varying dynamic graph convolution instead of static graph convolution (ablation) | Graph-ODE · MuPoTS-3D | MPJPE: 0.4066 -> 0.3706 | Improved |
| use hierarchical structure instead of flat structure (ablation) | DG-ODE · MuPoTS-3D | MPJPE: 0.3706 -> 0.3038 | Improved |
| use hierarchical structure instead of flat structure (ablation) | ODE-RNN · MuPoTS-3D | MPJPE: 0.4396 -> 0.4938 | Hurt |
| use hierarchical structure instead of flat structure (ablation) | DG-ODE · MuPoTS-3D | parameters: 10649 -> 2643 | Improved |
| use hierarchical structure instead of flat structure (ablation) | DG-ODE · MuPoTS-3D | memory: 0.76M -> 0.66M | Improved |
| use HDG-ODE instead of baselines under lower observable ratios | Human3.6M · Human3.6M (pt=0.6, ps=0.4) | MPJPE: 0.3826 -> 0.3543 (vs ODE-RNN) | Improved |
Closest papers by meaning
- HiPART: Hierarchical Pose AutoRegressive Transformer for Occluded 3D Human Pose Estimation (CVPR 2025 · similarity 0.89)
- Learning Dynamic Relationships for 3D Human Motion Prediction (CVPR 2020 · similarity 0.89)
- Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking (CVPR 2021 · similarity 0.89)
- $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.88)
- MotionMap: Representing Multimodality in Human Pose Forecasting (CVPR 2025 · similarity 0.87)
- Disentangled Diffusion-Based 3D Human Pose Estimation with Hierarchical Spatial and Temporal Denoiser (AAAI 2024 · similarity 0.87)
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.
