Cargando…
An Effective Conversion of Visemes to Words for High-Performance Automatic Lipreading
As an alternative approach, viseme-based lipreading systems have demonstrated promising performance results in decoding videos of people uttering entire sentences. However, the overall performance of such systems has been significantly affected by the efficiency of the conversion of visemes to words...
Autores principales: | Fenghour, Souheil, Chen, Daqing, Guo, Kun, Li, Bo, Xiao, Perry |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8659639/ https://www.ncbi.nlm.nih.gov/pubmed/34883888 http://dx.doi.org/10.3390/s21237890 |
Ejemplares similares
-
Learning the Relative Dynamic Features for Word-Level Lipreading
por: Li, Hao, et al.
Publicado: (2022) -
VISEM-Tracking, a human spermatozoa tracking dataset
por: Thambawita, Vajira, et al.
Publicado: (2023) -
Research on the dynamic viseme of the lip shape based on facial motion capture technology
por: Zhu, Shengyin
Publicado: (2022) -
Visual Cortical Entrainment to Motion and Categorical Speech Features during Silent Lipreading
por: O’Sullivan, Aisling E., et al.
Publicado: (2017) -
Visual inputs decrease brain activity in frontal areas during silent lipreading
por: Plata Bello, Julio, et al.
Publicado: (2019)