3D Shape Variational Autoencoder Latent Disentanglement via Mini-Batch Feature Swapping for Bodies and Faces

11/24/2021
by   Simone Foti, et al.
33

Learning a disentangled, interpretable, and structured latent representation in 3D generative models of faces and bodies is still an open problem. The problem is particularly acute when control over identity features is required. In this paper, we propose an intuitive yet effective self-supervised approach to train a 3D shape variational autoencoder (VAE) which encourages a disentangled latent representation of identity features. Curating the mini-batch generation by swapping arbitrary features across different shapes allows to define a loss function leveraging known differences and similarities in the latent representations. Experimental results conducted on 3D meshes show that state-of-the-art methods for latent disentanglement are not able to disentangle identity features of faces and bodies. Our proposed method properly decouples the generation of such features while maintaining good representation and reconstruction capabilities.

READ FULL TEXT

page 1

page 5

page 12

page 13

page 14

page 15

page 16

page 17

research
02/23/2023

Causally Disentangled Generative Variational AutoEncoder

We propose a new supervised learning method for Variational AutoEncoder ...
research
09/06/2023

Self-Supervised Disentanglement of Harmonic and Rhythmic Features in Music Audio Signals

The aim of latent variable disentanglement is to infer the multiple info...
research
04/17/2019

Learning Interpretable Disentangled Representations using Adversarial VAEs

Learning Interpretable representation in medical applications is becomin...
research
02/24/2023

3D Generative Model Latent Disentanglement via Local Eigenprojection

Designing realistic digital humans is extremely complex. Most data-drive...
research
08/06/2021

GLASS: Geometric Latent Augmentation for Shape Spaces

We investigate the problem of training generative models on a very spars...
research
08/18/2019

Geometric Disentanglement for Generative Latent Shape Models

Representing 3D shape is a fundamental problem in artificial intelligenc...
research
11/21/2020

Use of Student's t-Distribution for the Latent Layer in a Coupled Variational Autoencoder

A Coupled Variational Autoencoder, which incorporates both a generalized...

Please sign up or login with your details

Forgot password? Click here to reset