Goto

Collaborating Authors

 deepshell


Review for NeurIPS paper: Deep Shells: Unsupervised Shape Correspondence with Optimal Transport

Neural Information Processing Systems

Additional Feedback: My score is borderline: I think this work is interesting and it push further the state-of-the-art of the matching methods. However, I think there are some lacks in the experimental section, and I cannot see any clear progress from the Smooth Shells method. I am really curious to read the authors rebutal. Further questions: 1) Would be possible to have a video of the shells registration at test time? Do you think it would be beneficial in some cases e.g.