AI Paper Atlas

Research map / 3D vision / 3D generation

PhysTwin: Physics-Informed Reconstruction and Simulation of Deformable Objects from Videos

Hanxiao Jiang, Hao-Yu Hsu, Kaifeng Zhang, Hsin-Ni Yu, Shenlong Wang, Yunzhu Li

ICCV 2025 · 3D vision › 3D generation · approach: Learning physics from video · 51 citations (OpenAlex)

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 changedWhereMeasuredEffect
add hierarchical sparse-to-dense optimization (zero-order then first-order) over zero-order only (ablation)PhysTwin · Future PredictionCD: 0.014 -> 0.012Improved
add hierarchical sparse-to-dense optimization over zero-order only (ablation)PhysTwin · Reconstruction & ResimulationIoU %: 80.2 -> 84.4Improved
add zero-order sparse optimization of topology on top of first-order only (ablation)PhysTwin · Future PredictionCD: 0.019 -> 0.012Improved
add zero-order sparse optimization on top of first-order only (ablation)PhysTwin · Reconstruction & ResimulationIoU %: 82.7 -> 84.4Improved
add first-order dense stiffness optimization on top of zero-order only (ablation)PhysTwin · Future PredictionIoU %: 69.2 -> 72.5Improved
use PhysTwin instead of GS-Dynamicsreconstruction and resimulation · Reconstruction & ResimulationIoU %: 72.1 -> 84.4Improved
use PhysTwin instead of Spring-Gausfuture prediction · Future PredictionCD: 0.062 -> 0.012Improved
use PhysTwin instead of GS-Dynamicsfuture prediction · Future PredictionPSNR: 22.540 -> 25.617Improved
use PhysTwin instead of GS-Dynamicsgeneralization to unseen interactions · Generalization to Unseen InteractionsCD: 0.029 -> 0.013Improved
use PhysTwin instead of GS-Dynamicsgeneralization to unseen interactions · Generalization to Unseen InteractionsPSNR: 25.053 -> 26.199Improved
train GS-Dynamics with 30 data points instead of 1GS-Dynamics · rope double-hand stretching and foldingperformanceNo clear effect
optimize physics then use tracking after physics-constrained optimization instead of raw CoTracker3 tracksPhysTwin · trackingtracking qualityImproved

Later papers that compared with this method

2 later accepted papers report PhysTwin as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy