Disentangling Geometric Deformation Spaces in Generative Latent Shape Models

TL;DR

This paper advances a generative model for 3D shapes that disentangles geometric deformations into interpretable factors like pose and shape, enabling pose transfer and retrieval without supervision.

Contribution

It introduces improved handling of rotational invariance and a diffeomorphic flow network to better factorize and interpret 3D shape deformations in an unsupervised manner.

Findings

Enhanced rotation invariance in shape representations

Improved intrinsic-extrinsic shape factorization

Effective unsupervised pose transfer and retrieval

Abstract

A complete representation of 3D objects requires characterizing the space of deformations in an interpretable manner, from articulations of a single instance to changes in shape across categories. In this work, we improve on a prior generative model of geometric disentanglement for 3D shapes, wherein the space of object geometry is factorized into rigid orientation, non-rigid pose, and intrinsic shape. The resulting model can be trained from raw 3D shapes, without correspondences, labels, or even rigid alignment, using a combination of classical spectral geometry and probabilistic disentanglement of a structured latent representation space. Our improvements include more sophisticated handling of rotational invariance and the use of a diffeomorphic flow network to bridge latent and spectral space. The geometric structuring of the latent space imparts an interpretable characterization of the deformation space of an object. Furthermore, it enables tasks like pose transfer and pose-aware retrieval without requiring supervision. We evaluate our model on its generative modelling, representation learning, and disentanglement performance, showing improved rotation invariance and intrinsic-extrinsic factorization quality over the prior model.