LipGen: Viseme-Guided Lip Video Generation for Enhancing Visual Speech Recognition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Hao, Bowen, Zhou, Dongliang, Li, Xiaojie, Zhang, Xingyu, Xie, Liang, Wu, Jianlong, Yin, Erwei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929665044643840
author Hao, Bowen
Zhou, Dongliang
Li, Xiaojie
Zhang, Xingyu
Xie, Liang
Wu, Jianlong
Yin, Erwei
author_facet Hao, Bowen
Zhou, Dongliang
Li, Xiaojie
Zhang, Xingyu
Xie, Liang
Wu, Jianlong
Yin, Erwei
contents Visual speech recognition (VSR), commonly known as lip reading, has garnered significant attention due to its wide-ranging practical applications. The advent of deep learning techniques and advancements in hardware capabilities have significantly enhanced the performance of lip reading models. Despite these advancements, existing datasets predominantly feature stable video recordings with limited variability in lip movements. This limitation results in models that are highly sensitive to variations encountered in real-world scenarios. To address this issue, we propose a novel framework, LipGen, which aims to improve model robustness by leveraging speech-driven synthetic visual data, thereby mitigating the constraints of current datasets. Additionally, we introduce an auxiliary task that incorporates viseme classification alongside attention mechanisms. This approach facilitates the efficient integration of temporal information, directing the model's focus toward the relevant segments of speech, thereby enhancing discriminative capabilities. Our method demonstrates superior performance compared to the current state-of-the-art on the lip reading in the wild (LRW) dataset and exhibits even more pronounced advantages under challenging conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2501_04204
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LipGen: Viseme-Guided Lip Video Generation for Enhancing Visual Speech Recognition
Hao, Bowen
Zhou, Dongliang
Li, Xiaojie
Zhang, Xingyu
Xie, Liang
Wu, Jianlong
Yin, Erwei
Computer Vision and Pattern Recognition
Multimedia
Visual speech recognition (VSR), commonly known as lip reading, has garnered significant attention due to its wide-ranging practical applications. The advent of deep learning techniques and advancements in hardware capabilities have significantly enhanced the performance of lip reading models. Despite these advancements, existing datasets predominantly feature stable video recordings with limited variability in lip movements. This limitation results in models that are highly sensitive to variations encountered in real-world scenarios. To address this issue, we propose a novel framework, LipGen, which aims to improve model robustness by leveraging speech-driven synthetic visual data, thereby mitigating the constraints of current datasets. Additionally, we introduce an auxiliary task that incorporates viseme classification alongside attention mechanisms. This approach facilitates the efficient integration of temporal information, directing the model's focus toward the relevant segments of speech, thereby enhancing discriminative capabilities. Our method demonstrates superior performance compared to the current state-of-the-art on the lip reading in the wild (LRW) dataset and exhibits even more pronounced advantages under challenging conditions.
title LipGen: Viseme-Guided Lip Video Generation for Enhancing Visual Speech Recognition
topic Computer Vision and Pattern Recognition
Multimedia
url https://arxiv.org/abs/2501.04204