NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:4751
Title:PRNet: Self-Supervised Learning for Partial-to-Partial Registration

This paper was well-received by the reviewers, who noted that it has a novel method with significant improvement compared to previous methods. The reviewers all requested that more analysis be performed to understand the contribution and limitations of different components of the method. It would also be helpful to show comparison to other baselines such as: Deep Part Induction from Articulated Object Pairs PointNetLK