AU-Blendshape for Fine-grained Stylized 3D Facial Expression Manipulation

TL;DR

This paper introduces AUBlendSet, a new 3D facial dataset based on AU-Blendshape representation, and AUBlendNet, a neural network for stylized 3D facial expression manipulation across identities, validated through various facial animation tasks.

Contribution

The paper presents the first dataset and network for continuous 3D facial expression manipulation using facial AUs across multiple identities.

Findings

AUBlendSet enables detailed AU annotation across 500 identities.

AUBlendNet effectively predicts style-specific AU-Blendshape basis vectors.

The approach improves stylized facial expression manipulation and emotion recognition data augmentation.

Abstract

While 3D facial animation has made impressive progress, challenges still exist in realizing fine-grained stylized 3D facial expression manipulation due to the lack of appropriate datasets. In this paper, we introduce the AUBlendSet, a 3D facial dataset based on AU-Blendshape representation for fine-grained facial expression manipulation across identities. AUBlendSet is a blendshape data collection based on 32 standard facial action units (AUs) across 500 identities, along with an additional set of facial postures annotated with detailed AUs. Based on AUBlendSet, we propose AUBlendNet to learn AU-Blendshape basis vectors for different character styles. AUBlendNet predicts, in parallel, the AU-Blendshape basis vectors of the corresponding style for a given identity mesh, thereby achieving stylized 3D emotional facial manipulation. We comprehensively validate the effectiveness of AUBlendSet and AUBlendNet through tasks such as stylized facial expression manipulation, speech-driven emotional facial animation, and emotion recognition data augmentation. Through a series of qualitative and quantitative experiments, we demonstrate the potential and importance of AUBlendSet and AUBlendNet in 3D facial animation tasks. To the best of our knowledge, AUBlendSet is the first dataset, and AUBlendNet is the first network for continuous 3D facial expression manipulation for any identity through facial AUs. Our source code is available at https://github.com/wslh852/AUBlendNet.git.