DexDiffuser
YuLiHN
A grasp sampler, evaluator and refinement pipeline conditioned on partial object point clouds.
What you can explore
Explore grasp proposals when the full object surface is not observable.
Before you use it
This is the grasp-generation project by YuLiHN, separate from DexHandDiff. Sensor quality, object coverage and embodiment affect transfer.
Original contributors
DexDiffuser authors and repository contributors
License / access: Code: MIT · data, models and dependencies may have separate terms
Keep exploring
DexHandDiff
Interaction-aware diffusion planning for dexterous manipulation.
DexGraspAnything
Physics-aware dexterous grasp generation across varied objects.
GenDexGrasp
Contact-map-based grasp synthesis designed to generalize across robotic hand embodiments.
Editorial reference · checked 2026-09-22