Make Your Actor Talk: Generalizable and High-Fidelity Lip Sync with Motion and Appearance Disentanglement

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Runyi, He, Tianyu, Zhang, Ailing, Wang, Yuchi, Guo, Junliang, Tan, Xu, Liu, Chang, Chen, Jie, Bian, Jiang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909225151627264
author Yu, Runyi
He, Tianyu
Zhang, Ailing
Wang, Yuchi
Guo, Junliang
Tan, Xu
Liu, Chang
Chen, Jie
Bian, Jiang
author_facet Yu, Runyi
He, Tianyu
Zhang, Ailing
Wang, Yuchi
Guo, Junliang
Tan, Xu
Liu, Chang
Chen, Jie
Bian, Jiang
contents We aim to edit the lip movements in talking video according to the given speech while preserving the personal identity and visual details. The task can be decomposed into two sub-problems: (1) speech-driven lip motion generation and (2) visual appearance synthesis. Current solutions handle the two sub-problems within a single generative model, resulting in a challenging trade-off between lip-sync quality and visual details preservation. Instead, we propose to disentangle the motion and appearance, and then generate them one by one with a speech-to-motion diffusion model and a motion-conditioned appearance generation model. However, there still remain challenges in each stage, such as motion-aware identity preservation in (1) and visual details preservation in (2). Therefore, to preserve personal identity, we adopt landmarks to represent the motion, and further employ a landmark-based identity loss. To capture motion-agnostic visual details, we use separate encoders to encode the lip, non-lip appearance and motion, and then integrate them with a learned fusion module. We train MyTalk on a large-scale and diverse dataset. Experiments show that our method generalizes well to the unknown, even out-of-domain person, in terms of both lip sync and visual detail preservation. We encourage the readers to watch the videos on our project page (https://Ingrid789.github.io/MyTalk/).
format Preprint
id arxiv_https___arxiv_org_abs_2406_08096
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Make Your Actor Talk: Generalizable and High-Fidelity Lip Sync with Motion and Appearance Disentanglement
Yu, Runyi
He, Tianyu
Zhang, Ailing
Wang, Yuchi
Guo, Junliang
Tan, Xu
Liu, Chang
Chen, Jie
Bian, Jiang
Computer Vision and Pattern Recognition
We aim to edit the lip movements in talking video according to the given speech while preserving the personal identity and visual details. The task can be decomposed into two sub-problems: (1) speech-driven lip motion generation and (2) visual appearance synthesis. Current solutions handle the two sub-problems within a single generative model, resulting in a challenging trade-off between lip-sync quality and visual details preservation. Instead, we propose to disentangle the motion and appearance, and then generate them one by one with a speech-to-motion diffusion model and a motion-conditioned appearance generation model. However, there still remain challenges in each stage, such as motion-aware identity preservation in (1) and visual details preservation in (2). Therefore, to preserve personal identity, we adopt landmarks to represent the motion, and further employ a landmark-based identity loss. To capture motion-agnostic visual details, we use separate encoders to encode the lip, non-lip appearance and motion, and then integrate them with a learned fusion module. We train MyTalk on a large-scale and diverse dataset. Experiments show that our method generalizes well to the unknown, even out-of-domain person, in terms of both lip sync and visual detail preservation. We encourage the readers to watch the videos on our project page (https://Ingrid789.github.io/MyTalk/).
title Make Your Actor Talk: Generalizable and High-Fidelity Lip Sync with Motion and Appearance Disentanglement
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.08096