3D Generative Model Latent Disentanglement via Local Eigenprojection

TL;DR

This paper introduces a spectral geometry-based loss function for neural network models of 3D meshes, enhancing local attribute control and disentanglement in digital human generation.

Contribution

A novel local eigenprojection loss function that improves latent disentanglement in 3D mesh generative models without sacrificing generation quality.

Findings

Enhanced disentanglement over state-of-the-art methods

Maintained good generation quality

Training times comparable to standard models

Abstract

Designing realistic digital humans is extremely complex. Most data-driven generative models used to simplify the creation of their underlying geometric shape do not offer control over the generation of local shape attributes. In this paper, we overcome this limitation by introducing a novel loss function grounded in spectral geometry and applicable to different neural-network-based generative models of 3D head and body meshes. Encouraging the latent variables of mesh variational autoencoders (VAEs) or generative adversarial networks (GANs) to follow the local eigenprojections of identity attributes, we improve latent disentanglement and properly decouple the attribute creation. Experimental results show that our local eigenprojection disentangled (LED) models not only offer improved disentanglement with respect to the state-of-the-art, but also maintain good generation capabilities with training times comparable to the vanilla implementations of the models.