Controllable 3D Generative Adversarial Face Model via Disentangling Shape and Appearance

TL;DR

This paper introduces a novel 3D face generative model that disentangles identity and expression, enabling precise control over facial expressions while maintaining individual identity, advancing capabilities in virtual avatar creation and synthetic data generation.

Contribution

It proposes a new 3D face generative framework using supervised auto-encoders and GANs to decouple shape and appearance, allowing fine-grained expression control.

Findings

Successfully decouples identity and expression in 3D face generation

Enables granular control over facial expressions

Preserves identity while manipulating expressions

Abstract

3D face modeling has been an active area of research in computer vision and computer graphics, fueling applications ranging from facial expression transfer in virtual avatars to synthetic data generation. Existing 3D deep learning generative models (e.g., VAE, GANs) allow generating compact face representations (both shape and texture) that can model non-linearities in the shape and appearance space (e.g., scatter effects, specularities, etc.). However, they lack the capability to control the generation of subtle expressions. This paper proposes a new 3D face generative model that can decouple identity and expression and provides granular control over expressions. In particular, we propose using a pair of supervised auto-encoder and generative adversarial networks to produce high-quality 3D faces, both in terms of appearance and shape. Experimental results in the generation of 3D faces learned with holistic expression labels, or Action Unit labels, show how we can decouple identity and expression; gaining fine-control over expressions while preserving identity.