Research map / 3D vision / 3D generation
PhysHO: Physics-Based Dynamic 3D Gaussian Human and Object from Monocular Video
Suyi Jiang, Gim Hee Lee
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 physics-aware fine-tuning of canonical Gaussians (ablation) | PhysHO · Spin stage rendering | PSNR: 25.42 -> 27.30 | Improved |
| add per-particle feature vectors le, lp to the neural constitutive laws (ablation) | PhysHO · dynamic reconstruction | PSNR: 23.54 -> 24.03 | Improved |
| add neural residual constitutive laws on top of expert models (ablation) | PhysHO · dynamic reconstruction | PSNR: 22.26 -> 24.03 | Improved |
| add neural residual constitutive laws on top of expert models (ablation) | PhysHO · dynamic reconstruction | IoU: 0.8289 -> 0.8845 | Improved |
| add neural residual constitutive laws on top of expert models (ablation) | PhysHO · dynamic reconstruction | LPIPS: 0.0664 -> 0.0652 | No clear effect |
| add 3D flow supervision (ablation) | PhysHO · neural residual constitutive laws training | training stability | Improved |
| use PhysHO instead of GART for reconstruction LPIPS | PhysHO · Square Pillow reconstruction (full sequence) | LPIPS: 0.1282 -> 0.1079 | Improved |
| use PhysHO instead of 4D Gaussian for reconstruction PSNR | PhysHO · Square Pillow reconstruction (full sequence) | PSNR: 27.64 -> 21.69 | Hurt |
| use PhysHO instead of GART for future prediction | PhysHO · C-shape Pillow #1 prediction (full sequence) | PSNR: 21.69 -> 21.12 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GART — GART: Gaussian Articulated Template Models (CVPR 2024) · better on 8 of 9 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
- PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation (NeurIPS 2025 · similarity 0.90)
- PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videos (CVPR 2024 · similarity 0.89)
- DeformMaster: An Interactive Physics-Neural World Model for Deformable Objects from Videos (NeurIPS 2026 · similarity 0.88)
- OmniPhysGS: 3D Constitutive Gaussians for General Physics-Based Dynamics Generation (ICLR 2025 · similarity 0.88)
- MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos (NeurIPS 2026 · similarity 0.87)
- PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics (CVPR 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.
