AI Paper Atlas

Research map / 3D vision / 3D generation

GeoDexGrasp: Geometry-aware Generation for Data-efficient and Physics-plausible Dexterous Grasping

Bing Han, Weiyuan Liu, Changlong Zhang, Chenxi Wang, Zhibin Zhao, Zhi Zhai

CVPR 2026 · 3D vision › 3D generation · approach: Hand-object interaction generation

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
decouple root rotation (SO(3)) from finger joints and translation (Euclidean) (ablation)GeoDexGrasp · DexGRABsuccess rate (%): 41.8 -> 56.1Improved
add pose and interaction-direction representations (ablation)GeoDexGrasp (decoupled) · DexGRABsuccess rate (%): 56.1 -> 58.6Improved
add shape and scale representations (ablation)GeoDexGrasp (decoupled) · DexGRABpenetration depth (mm): 31.5 -> 19.9Improved
add self-supervised disentanglement pretraining of the SIM(3) encoder (ablation)GeoDexGrasp · DexGRABsuccess rate (%): 63.2 -> 67.5Improved
use geometry-aware SIM(3)-equivariant representations instead of DexGraspAnythingdexterous grasp generation · Five datasets averagepenetration depth (mm): 21.2 -> 13.6Improved
use geometry-aware representations with far fewer parametersGeoDexGrasp vs DexGraspAnything · Five datasets averagesuccess rate (%): 58.5 -> 60.1 with 159.7M -> 28.7M parametersNo clear effect
train on reduced data (25% of grasps per object)GeoDexGrasp vs UGG and DexGraspAnything · UniDexGrasp and DexGraspNetsuccess rateImproved
geometry-aware model on out-of-distribution object sizesGeoDexGrasp · Size generalization (Cube, Cylinder, Torus)success rate (%): 56.3 -> 68.8 on large cubeImproved
geometry-aware model on topologically complex Toy horseGeoDexGrasp · Size generalization, Toy horsesuccess rate (%): 9.4 (large)No clear effect
explicitly extract scale factorGeoDexGrasp · DexGRABsuccess rateNo clear effect
geometry-aware model on RealDexGeoDexGrasp · RealDexsuccess rate (%): 44.8 (DexGraspAnything) vs 38.3Hurt

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