NeuroMorph: Unsupervised Shape Interpolation and Correspondence in One Go

TL;DR

NeuroMorph is an unsupervised neural network that efficiently computes shape interpolation and correspondence between 3D shapes in a single forward pass, leveraging geometric priors for high accuracy.

Contribution

It introduces a novel architecture combining graph convolutions and global pooling for unsupervised shape correspondence and interpolation in one go.

Findings

Achieves state-of-the-art results on shape correspondence benchmarks.

Handles non-isometric shape pairs from different categories.

Operates efficiently with a single feed-forward pass.

Abstract

We present NeuroMorph, a new neural network architecture that takes as input two 3D shapes and produces in one go, i.e. in a single feed forward pass, a smooth interpolation and point-to-point correspondences between them. The interpolation, expressed as a deformation field, changes the pose of the source shape to resemble the target, but leaves the object identity unchanged. NeuroMorph uses an elegant architecture combining graph convolutions with global feature pooling to extract local features. During training, the model is incentivized to create realistic deformations by approximating geodesics on the underlying shape space manifold. This strong geometric prior allows to train our model end-to-end and in a fully unsupervised manner without requiring any manual correspondence annotations. NeuroMorph works well for a large variety of input shapes, including non-isometric pairs from different object categories. It obtains state-of-the-art results for both shape correspondence and interpolation tasks, matching or surpassing the performance of recent unsupervised and supervised methods on multiple benchmarks.