NeurIPS 2020

Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernels

Meta Review

All the reviewers voted for the paper to be accepted and praised the novelty of the proposed mesh autoencoder. They are however expecting the changes promised in the rebuttal to be included in the camera ready version of the paper. They also suggest stating clearly in the paper the limitations -- namely that connectivity is already solved in the training set, which limits applicability of the method.