Skip to content
Massachusetts Institute of Technology
  • on: Dec. 9, 2021
  • in: ICRA

Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulation

  • Anthony Simeonov *
  • Yilun Du *
  • Andrea Tagliasacchi
  • Alberto Rodriguez
  • Pulkit Agrawal
  • Vincent Sitzmann
*
shared first author
equal advising
@inproceedings{simeonovdu2021ndf,
    title = { Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulation },
    author = { Simeonov, Anthony and 
               Du, Yilun and 
               Tagliasacchi, Andrea and 
               Rodriguez, Alberto and 
               Agrawal, Pulkit and 
               Sitzmann, Vincent },
    year = { 2021 },
    booktitle = { ICRA },
}
  • Copy to Clipboard

We present Neural Descriptor Fields (NDFs), an object representation that encodes both points and relative poses between an object and a target — such as a robot gripper or a rack used for hanging — via category-level descriptors.

We employ this representation for object manipulation, where given a task demonstration, we want to repeat the same task on a new object instance from the same category. We propose to achieve this objective by searching (via optimization) for the pose whose descriptor matches that observed in the demonstration.

NDFs are conveniently trained in a self-supervised fashion via a 3D auto-encoding task that does not rely on expert-labeled keypoints. Further, NDFs are SE(3)-equivariant, guaranteeing performance that generalizes across all possible 3D object translations and rotations.

We demonstrate learning of manipulation tasks from few (5–10) demonstrations both in simulation and on a real robot. Our performance generalizes across both object instances and 6-DoF object poses, and significantly outperforms a recent baseline that relies on 2D descriptors.