Everybody's Talkin': Let Me Talk as You Want

TL;DR

This paper introduces a dynamic, end-to-end method for editing portrait videos by translating source audio into realistic facial movements, preserving original geometry and pose, and ensuring temporal coherence.

Contribution

It presents a novel approach that factorizes video into expression, geometry, and pose, using a recurrent network to map audio to expressions without person-specific training.

Findings

Achieves high realism in talking portrait videos

Maintains original video context and pose

Robust to variations in source audio

Abstract

We present a method to edit a target portrait footage by taking a sequence of audio as input to synthesize a photo-realistic video. This method is unique because it is highly dynamic. It does not assume a person-specific rendering network yet capable of translating arbitrary source audio into arbitrary video output. Instead of learning a highly heterogeneous and nonlinear mapping from audio to the video directly, we first factorize each target video frame into orthogonal parameter spaces, i.e., expression, geometry, and pose, via monocular 3D face reconstruction. Next, a recurrent network is introduced to translate source audio into expression parameters that are primarily related to the audio content. The audio-translated expression parameters are then used to synthesize a photo-realistic human subject in each video frame, with the movement of the mouth regions precisely mapped to the source audio. The geometry and pose parameters of the target human portrait are retained, therefore preserving the context of the original video footage. Finally, we introduce a novel video rendering network and a dynamic programming method to construct a temporally coherent and photo-realistic video. Extensive experiments demonstrate the superiority of our method over existing approaches. Our method is end-to-end learnable and robust to voice variations in the source audio.