Self-Supervised Adaptation of High-Fidelity Face Models for Monocular Performance Tracking

TL;DR

This paper introduces a self-supervised domain adaptation method that enables high-fidelity face models to be driven from monocular cellphone images without labeled data, overcoming domain and input data constraints.

Contribution

It presents a novel approach combining direct face model control from 2D images with self-supervised domain adaptation using frame texture consistency, eliminating the need for specialized input data or labeled target domain data.

Findings

Successfully drives high-fidelity face models from cellphone images

No labeled data needed for domain adaptation

Maintains realistic facial motion in uncontrolled environments

Abstract

Improvements in data-capture and face modeling techniques have enabled us to create high-fidelity realistic face models. However, driving these realistic face models requires special input data, e.g. 3D meshes and unwrapped textures. Also, these face models expect clean input data taken under controlled lab environments, which is very different from data collected in the wild. All these constraints make it challenging to use the high-fidelity models in tracking for commodity cameras. In this paper, we propose a self-supervised domain adaptation approach to enable the animation of high-fidelity face models from a commodity camera. Our approach first circumvents the requirement for special input data by training a new network that can directly drive a face model just from a single 2D image. Then, we overcome the domain mismatch between lab and uncontrolled environments by performing self-supervised domain adaptation based on "consecutive frame texture consistency" based on the assumption that the appearance of the face is consistent over consecutive frames, avoiding the necessity of modeling the new environment such as lighting or background. Experiments show that we are able to drive a high-fidelity face model to perform complex facial motion from a cellphone camera without requiring any labeled data from the new domain.