NeurIPS 2020

Dense Correspondences between Human Bodies via Learning Transformation Synchronization on Graphs

Meta Review

This submission proposes an approach to establishing dense correspondences between full and partial human body scans. It initially received four reviews with three positive scores (6,8,6,5). The reviewers appreciated strong performance, efficiency, novelty in the context of the given application and solid theoretical and empirical analysis. At the same time, reviewers would like to see the proposed method applied to other object categories with different topologies. The rebuttal addressed some of the remaining concerns, which resulted in an increase in scores to (6,8,6,6). For these reasons, the AC's recommendation is to accept this submission for presentation as a poster.