NeurIPS 2020

Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloud

Meta Review

Post-rebuttal, the reviews were borderline, but leaning positive. The reviewers had concerns about novelty, but thought the paper was a good practical contribution to an important and timely problem. The AC is inclined to agree with the positively-inclined reviewers. The AC urges authors to include experiments from the rebuttal as well as timing information in the camera-ready version of the paper, as this was important for the reviewers' discussions.