Research map / 3D vision / Human pose estimation
PPR: Physically Plausible Reconstruction from Monocular Videos
Gengshan Yang, Shuo Yang, John Z. Zhang, Zachary Manchester, Deva Ramanan
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 |
|---|---|---|---|
| replace BANMo control-point deformation with skeleton deformation | BANMo · AMA samba | Chamfer distance (cm): 11.8 -> 8.9 | Improved |
| same skeleton deformation swap | BANMo · AMA bouncing | Chamfer distance (cm): 12.8 -> 14.1 | Hurt |
| add differentiable physics simulation on top of skeleton model | BANMo +skel · AMA samba | Chamfer distance (cm): 8.9 -> 8.3 | Improved |
| use differentiable physics instead of ground-prior scale fitting (ablation) | PPR · AMA samba | contact (%): 26.3 -> 67.4 | Improved |
| use coordinate descent alternating rendering and physics instead of one cycle (ablation) | PPR · AMA samba | Chamfer distance (cm): 46.0 -> 8.3 | Improved |
| coordinate descent vs one cycle on foot skate (ablation) | PPR · AMA samba | foot skate (cm): 1.1 -> 1.2 | Mixed |
| use PD feedback control instead of open-loop control (ablation) | PPR · AMA samba | contact (%): 53.6 -> 67.4 | Improved |
| optimize mass and PD gains instead of freezing (ablation) | PPR · AMA samba | contact (%): 50.4 -> 67.4 | Improved |
| PPR vs HuMoR on foot skate | PPR · AMA bouncing | foot skate (cm): 2.6 -> 7.4 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- HuMoR — HuMoR: 3D Human Motion Model for Robust Pose Estimation (ICCV 2021) · better on 3 of 4 shared results
- BANMo — BANMo: Building Animatable 3D Neural Models From Many Casual Videos (CVPR 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Differentiable Dynamics for Articulated 3D Human Motion Reconstruction (CVPR 2022 · similarity 0.92)
- MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos (NeurIPS 2026 · similarity 0.90)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.87)
- REACTO: Reconstructing Articulated Objects from a Single Video (CVPR 2024 · similarity 0.87)
- Real2Sim in HOI: Toward Physically Plausible HOI Reconstruction from Monocular Videos (NeurIPS 2026 · similarity 0.86)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.86)
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.
