REFA: Real-time Egocentric Facial Animations for Virtual Reality

TL;DR

REFA is a real-time, non-intrusive system that captures egocentric facial expressions via embedded infrared cameras in VR headsets, enabling realistic virtual character animations without extensive calibration.

Contribution

We introduce a novel distillation-based machine learning approach trained on diverse synthetic and real data for real-time facial expression tracking in VR.

Findings

Achieved accurate real-time facial expression tracking in VR.

Collected a dataset of 18,000 subjects with a lightweight setup.

Developed a differentiable rendering pipeline for automatic label extraction.

Abstract

We present a novel system for real-time tracking of facial expressions using egocentric views captured from a set of infrared cameras embedded in a virtual reality (VR) headset. Our technology facilitates any user to accurately drive the facial expressions of virtual characters in a non-intrusive manner and without the need of a lengthy calibration step. At the core of our system is a distillation based approach to train a machine learning model on heterogeneous data and labels coming form multiple sources, \eg synthetic and real images. As part of our dataset, we collected 18k diverse subjects using a lightweight capture setup consisting of a mobile phone and a custom VR headset with extra cameras. To process this data, we developed a robust differentiable rendering pipeline enabling us to automatically extract facial expression labels. Our system opens up new avenues for communication and expression in virtual environments, with applications in video conferencing, gaming, entertainment, and remote collaboration.