Research map / 3D vision / Human pose estimation
Trajectory Optimization for Physics-Based Reconstruction of 3D Human Pose From Monocular Video
Check 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 trajectory-optimized dynamics | HUND+SO+GT · Human3.6M validation subset | Footskate (%): 64 -> 8 | Helped |
| add trajectory-optimized dynamics | HUND+SO+GT · Human3.6M validation subset | MPJPE-G: 135 -> 132 | Helped |
| add trajectory-optimized dynamics | HUND+SO+GT · Human3.6M validation subset | Velocity: 0.58 -> 0.27 | Helped |
| add trajectory-optimized dynamics | HUND+SO+GT · Human3.6M | MPJPE-G: 145 -> 143 | Helped |
| add trajectory-optimized dynamics | HUND+SO+GT · HumanEva-I | MPJPE-G: 208 -> 196 | Helped |
| add trajectory-optimized dynamics | HUND+SO+GT · AIST | MPJPE-G: 156 -> 154 | Helped |
| remove 2d reprojection loss (ablation) | HUND+SO+GT + Dynamics · Human3.6M validation subset | MPJPE-G: 132 -> 154 | Hurt |
| remove 3d joints loss (ablation) | HUND+SO+GT + Dynamics · Human3.6M validation subset | MPJPE-G: 132 -> 134 | Hurt |
| remove COM loss (ablation) | HUND+SO+GT + Dynamics · Human3.6M validation subset | MPJPE-G: 132 -> 149 | Hurt |
| remove COM and 3d joints losses (ablation) | HUND+SO+GT + Dynamics · Human3.6M validation subset | MPJPE-G: 132 -> 151 | Hurt |
| remove pose prior (ablation) | HUND+SO+GT + Dynamics · Human3.6M validation subset | MPJPE-G: 132 -> 138 | Hurt |
| add time-consistent body shape (ablation) | HUND · Human3.6M validation subset | MPJPE-G: 239 -> 233 | Helped |
Closest papers by meaning
- Optimal-state Dynamics Estimation for Physics-based Human Motion Capture from Videos (NeurIPS 2024 · similarity 0.91)
- MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos (NeurIPS 2026 · similarity 0.89)
- PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videos (CVPR 2024 · similarity 0.89)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.88)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.88)
- WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion (CVPR 2024 · 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.