Robust Egocentric Photo-realistic Facial Expression Transfer for Virtual Reality

TL;DR

This paper introduces a robust, multi-identity facial expression transfer method for VR avatars that generalizes well across different users and environmental conditions, reducing the need for extensive personalized data.

Contribution

The paper proposes an end-to-end multi-identity architecture with specialized augmentation that decouples facial expressions from nuisance factors, improving robustness and generalization in VR facial expression transfer.

Findings

Outperforms state-of-the-art person-specific models in challenging scenarios

Effectively generalizes to untrained subjects with minimal personalized data

Demonstrates robustness against headset variability, lighting, and appearance changes

Abstract

Social presence, the feeling of being there with a real person, will fuel the next generation of communication systems driven by digital humans in virtual reality (VR). The best 3D video-realistic VR avatars that minimize the uncanny effect rely on person-specific (PS) models. However, these PS models are time-consuming to build and are typically trained with limited data variability, which results in poor generalization and robustness. Major sources of variability that affects the accuracy of facial expression transfer algorithms include using different VR headsets (e.g., camera configuration, slop of the headset), facial appearance changes over time (e.g., beard, make-up), and environmental factors (e.g., lighting, backgrounds). This is a major drawback for the scalability of these models in VR. This paper makes progress in overcoming these limitations by proposing an end-to-end multi-identity architecture (MIA) trained with specialized augmentation strategies. MIA drives the shape component of the avatar from three cameras in the VR headset (two eyes, one mouth), in untrained subjects, using minimal personalized information (i.e., neutral 3D mesh shape). Similarly, if the PS texture decoder is available, MIA is able to drive the full avatar (shape+texture) robustly outperforming PS models in challenging scenarios. Our key contribution to improve robustness and generalization, is that our method implicitly decouples, in an unsupervised manner, the facial expression from nuisance factors (e.g., headset, environment, facial appearance). We demonstrate the superior performance and robustness of the proposed method versus state-of-the-art PS approaches in a variety of experiments.