Research map / 3D vision / Face and shape
ReconPhys: Reconstruct Appearance and Physical Attributes from Single Video
Boyuan Wang, Xiaofeng Wang, Yongkang Li, Zheng Zhu, Yifan Chang, Angen Ye, Guosheng Zhao, Chaojun Ni, Guan Huang, Yijie Ren, Yueqi Duan, Xingang Wang
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 per-scene optimization (Spring-Gaus) with feedforward physics predictor | Spring-Gaus · Future prediction (46 unseen objects) | PSNR: 13.27 -> 21.64 | Improved |
| Replace Spring-Gaus with feedforward ReconPhys | Spring-Gaus · Future prediction (46 unseen objects) | CD: 0.349 -> 0.004 | Improved |
| Replace Spring-Gaus with feedforward ReconPhys | Spring-Gaus · Dynamic reconstruction (46 unseen objects) | PSNR: 22.26 -> 33.84 | Improved |
| Replace 4DGS with ReconPhys | 4DGS · Dynamic reconstruction (46 unseen objects) | CD: 0.593 -> 0.001 | Improved |
| Replace 4DGS with ReconPhys | 4DGS · Dynamic reconstruction (46 unseen objects) | SSIM: 0.983 -> 0.953 | Hurt |
| Replace 4DGS with ReconPhys | 4DGS · Dynamic reconstruction (46 unseen objects) | PSNR: 30.33 -> 33.84 | Improved |
| Replace per-scene optimization with feedforward inference | Spring-Gaus · Inference time | time per object: >1h -> <1s | Improved |
| Replace Spring-Gaus with ReconPhys | Spring-Gaus · Physical attribute estimation | MAE of stiffness: 827.67 -> 297.3 | Improved |
| Replace Spring-Gaus with ReconPhys | Spring-Gaus · Physical attribute estimation | MAE of mass: 2.276 -> 1.337 | Improved |
| Replace Spring-Gaus with ReconPhys | Spring-Gaus · Physical attribute estimation | MAE of friction: 1.082 -> 1.508 | Hurt |
| Replace Spring-Gaus with ReconPhys | Spring-Gaus · Physical disentanglement (Cylinder-1) | PSNR: 20.89 -> 36.86 | Improved |
Closest papers by meaning
- Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation (CVPR 2025 · similarity 0.87)
- ReconFusion: 3D Reconstruction with Diffusion Priors (CVPR 2024 · similarity 0.86)
- MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos (NeurIPS 2026 · similarity 0.86)
- OmniPhysGS: 3D Constitutive Gaussians for General Physics-Based Dynamics Generation (ICLR 2025 · similarity 0.86)
- ReCon-GS: Continuum-Preserved Guassian Streaming for Fast and Compact Reconstruction of Dynamic Scenes (NeurIPS 2025 · similarity 0.86)
- VideoPhy: Evaluating Physical Commonsense for Video Generation (ICLR 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.
