AI Paper Atlas

Research map / 3D vision / 3D generation

PhysHanDI: Physics-Based Reconstruction of Hand-Deformable Object Interactions

Jihyun Lee, Changmin Lee, Donghwan Kim, Tae-Kyun Kim

ICML 2026 · 3D vision › 3D generation · approach: Hand-object interaction generation · 0 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 inverse-physics hand refinement (ablation)PhysHanDI · PhysTwin-full, single-view future predictionChamfer distance: 42.8 -> 33.5Improved
add inverse-physics hand refinement (ablation)PhysHanDI · PhysTwin-full, single-view future predictionHand CD: 7.57 -> 7.17Improved
drive simulation with dense MANO hand vertices instead of ~30 sparse control pointsPhysTwin · PhysTwin-dense, reconstruction & resimulationCDdyn (mm): 10.78 -> 8.32Improved
drive simulation with dense MANO hand vertices instead of sparse control pointsPhysTwin · PhysTwin-dense, future predictionCDdyn (mm): 16.32 -> 14.35Improved
drive simulation with dense MANO hand vertices instead of sparse control pointsPhysTwin · DenseHDI, future predictionPSNR: 19.76 -> 19.75No clear effect
use dense hand reconstruction under perturbed tracking signalsPhysTwin · PhysTwin-dense, perturbed trackingCD (mm): 9.60 -> 5.56Improved
use dense hand reconstruction under perturbed controllerPhysTwin · PhysTwin-full, perturbed controllerTrack Err.: PhysTwin 1.25 vs ours 1.29Hurt
use dense hand points for contacts instead of sparse pointsspring-mass topology · PhysTwin-denseRRD virtual springs: 2.63 -> 0.35Improved
increase initial max connected nodes dmax to 50 (with delta 0.002) (ablation)PhysHanDI zero-order optimization · double_stretch_sloth sequenceCD: 4.44 -> 6.15Hurt
set initial connection radius delta to 0.001 instead of 0.002 (ablation)PhysHanDI zero-order optimization · double_stretch_sloth sequenceCD: 4.44 -> 6.81Hurt
add hand reconstruction and refinement stagesPhysHanDI pipeline · per-frame runtimeseconds per frame: object recon zero-order 21.22 -> 17.63 faster, hand recon adds 27.38Hurt

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.

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