UC Berkeley Releases Massive Dex-Net 2.0 Dataset

A dataset of 6.7 million robust point clouds and grasps can train your neural network to reliably pick up objects

5 min read

ABB YuMi arm grasps objects using Dex-Net 2.0 from UC Berkeley's AUTOLAB
A dataset of 6.7 million robust point clouds and grasps can train your neural network to reliably pick up objects.
Image: UC Berkeley/AUTOLAB