HO-3D Dataset Updates v2/v3 Annotations for Occluded Scenes

YuXiang_IRVL · x · 2026-08-16

HO-3D is a dataset for hand pose estimation during object interaction with severe occlusions. Ground-truth annotations for v2 and v3 have been released with evaluation scripts.

Original post →

More from Embodied

Embodied channel →