# SpatialDex: Pre-training 3D-Grounded Bimanual Dexterous Manipulation from RGB-Only Human Videos

- Source: arXiv cs.RO
- Published: 2026-10-22T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/febovrozipw6twaesdxl3mcmn
- Original: https://arxiv.org/abs/2610.29716

## Summary

SpatialDex is a framework that learns 3D-grounded bimanual dexterous manipulation from RGB-only human videos through reconstruction-augmented visuomotor pre-training, learning human behavior together with its spatial context. It reconstructs scene geometry and estimates hand motion from each video, aligns the two at a video-level scale using the hand-size prior of 3D hand estimates, and uses the reconstructed geometry as both a policy input and a prediction target.