A Novel Combination Of 3D CNNs And Recurrent Neural Networks for Sign Language to Text Conversion

Mehta, Pritish (2022) A Novel Combination Of 3D CNNs And Recurrent Neural Networks for Sign Language to Text Conversion. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (453kB) | Preview

Abstract

Sign Language Translation has recently achieved significant accomplishments, raising hopes for improved communication with the Deaf. The primary language of the Deaf population is now American Sign Language (ASL), which is conveyed via body language and understood through eye contact. The main objective of this study is to construct a deep learning-based automatic translation system that can translate ASL to English text. The WLASL dataset is used for the experiment. For sign-to-text translation, this study uses the CNN, GRU, CNN+LSTM, and planned 3D-CNN+LSTM networks. PCA is often employed as a pre-processing step to facilitate the identification of hands. The accuracy of the findings from the suggested model, which is better than the accuracy of the other models utilized in this study, was 83.33% at the end.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science P Language and Literature > P Philology. Linguistics > Semiotics > Language. Linguistic theory > Gesture. Sign language
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Tamara Malone
Date Deposited:	23 Feb 2023 12:07
Last Modified:	02 Mar 2023 09:22
URI:	https://norma.ncirl.ie/id/eprint/6226

Actions (login required)

View Item