UniSync: A Unified Framework for Audio-Visual Synchronization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Feng, Tao, Xie, Yifan, Guan, Xun, Song, Jiyuan, Liu, Zhou, Ma, Fei, Yu, Fei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916658209095680
author Feng, Tao
Xie, Yifan
Guan, Xun
Song, Jiyuan
Liu, Zhou
Ma, Fei
Yu, Fei
author_facet Feng, Tao
Xie, Yifan
Guan, Xun
Song, Jiyuan
Liu, Zhou
Ma, Fei
Yu, Fei
contents Precise audio-visual synchronization in speech videos is crucial for content quality and viewer comprehension. Existing methods have made significant strides in addressing this challenge through rule-based approaches and end-to-end learning techniques. However, these methods often rely on limited audio-visual representations and suboptimal learning strategies, potentially constraining their effectiveness in more complex scenarios. To address these limitations, we present UniSync, a novel approach for evaluating audio-visual synchronization using embedding similarities. UniSync offers broad compatibility with various audio representations (e.g., Mel spectrograms, HuBERT) and visual representations (e.g., RGB images, face parsing maps, facial landmarks, 3DMM), effectively handling their significant dimensional differences. We enhance the contrastive learning framework with a margin-based loss component and cross-speaker unsynchronized pairs, improving discriminative capabilities. UniSync outperforms existing methods on standard datasets and demonstrates versatility across diverse audio-visual representations. Its integration into talking face generation frameworks enhances synchronization quality in both natural and AI-generated content.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16357
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UniSync: A Unified Framework for Audio-Visual Synchronization
Feng, Tao
Xie, Yifan
Guan, Xun
Song, Jiyuan
Liu, Zhou
Ma, Fei
Yu, Fei
Computer Vision and Pattern Recognition
Sound
Audio and Speech Processing
Precise audio-visual synchronization in speech videos is crucial for content quality and viewer comprehension. Existing methods have made significant strides in addressing this challenge through rule-based approaches and end-to-end learning techniques. However, these methods often rely on limited audio-visual representations and suboptimal learning strategies, potentially constraining their effectiveness in more complex scenarios. To address these limitations, we present UniSync, a novel approach for evaluating audio-visual synchronization using embedding similarities. UniSync offers broad compatibility with various audio representations (e.g., Mel spectrograms, HuBERT) and visual representations (e.g., RGB images, face parsing maps, facial landmarks, 3DMM), effectively handling their significant dimensional differences. We enhance the contrastive learning framework with a margin-based loss component and cross-speaker unsynchronized pairs, improving discriminative capabilities. UniSync outperforms existing methods on standard datasets and demonstrates versatility across diverse audio-visual representations. Its integration into talking face generation frameworks enhances synchronization quality in both natural and AI-generated content.
title UniSync: A Unified Framework for Audio-Visual Synchronization
topic Computer Vision and Pattern Recognition
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2503.16357