The computer graphics and vision communities have dedicated long standing efforts in
building computerized tools for reconstructing, tracking, and analyzing human faces based …

Visual speech, referring to the visual domain of speech, has attracted increasing attention
due to its wide applications, such as public security, medical treatment, military defense, and …

Speech-driven 3D facial animation has been widely studied, yet there is still a gap to
achieving realism and vividness due to the highly ill-posed nature and scarcity of audio …

Generating high-fidelity talking head video by fitting with the input audio sequence is a
challenging problem that receives considerable attentions recently. In this paper, we …

One-shot talking face generation should synthesize high visual quality facial videos with
reasonable animations of expression and head pose, and just utilize arbitrary driving audio …

One-shot talking face generation aims at synthesizing a high-quality talking face video from
an arbitrary portrait image, driven by a video or an audio segment. In this work, we provide a …

Talking face generation has historically struggled to produce head movements and natural
facial expressions without guidance from additional reference videos. Recent developments …

To the best of our knowledge, we first present a live system that generates personalized
photorealistic talking-head animation only driven by audio signals at over 30 fps. Our system …

We present a method that generates expressive talking-head videos from a single facial
image with audio as the only input. In contrast to previous attempts to learn direct mappings …

This paper presents a generic method for generating full facial 3D animation from speech.
Existing approaches to audio-driven facial animation exhibit uncanny or static upper face …