FaceTuneGAN: Face Autoencoder for Convolutional Expression Transfer Using Neural Generative Adversarial Networks

TL;DR

FaceTuneGAN introduces a novel 3D face model that separately encodes identity and expression, enabling improved expression transfer and face neutralization through a neural network adapted from 2D image translation methods.

Contribution

It is the first adaptation of image-to-image translation networks to 3D face geometry, enhancing identity decomposition and expression transfer capabilities.

Findings

Outperforms state-of-the-art in identity decomposition and face neutralization.

Predicts blendshapes closer to ground-truth with fewer artifacts.

Effective on large face scan datasets.

Abstract

In this paper, we present FaceTuneGAN, a new 3D face model representation decomposing and encoding separately facial identity and facial expression. We propose a first adaptation of image-to-image translation networks, that have successfully been used in the 2D domain, to 3D face geometry. Leveraging recently released large face scan databases, a neural network has been trained to decouple factors of variations with a better knowledge of the face, enabling facial expressions transfer and neutralization of expressive faces. Specifically, we design an adversarial architecture adapting the base architecture of FUNIT and using SpiralNet++ for our convolutional and sampling operations. Using two publicly available datasets (FaceScape and CoMA), FaceTuneGAN has a better identity decomposition and face neutralization than state-of-the-art techniques. It also outperforms classical deformation transfer approach by predicting blendshapes closer to ground-truth data and with less of undesired artifacts due to too different facial morphologies between source and target.