NeurIPS 2020

Learning to Orient Surfaces by Self-supervised Spherical CNNs

Meta Review

This paper received overall positive ratings from three reviewers. However, the submission also received slightly conflicting comments in almost all aspects : novelty, writing and experiments. Some valid negative points concerning the empirical evaluation were raised by R1, and those were weighted amongst all reviewers during the discussion period. Taking all this into consideration, we decided on an acceptance as poster. I strongly encourage the authors to take the reviewer's feedback into account when preparing the camera-ready version, notably those concerning clarity and writing.