Proceedings of the International Conference on Emerging Intelligent Systems for Sustainable Development (ICEIS 2024)

<Previous Article In Volume
Next Article In Volume>

An LSTM-based System for Dynamic Arabic Sign Language Recognition

Authors
Slimane Oulad-Naoui1, Habiba Ben-Abderrahmane2, *, Assia Chagha1, Abderrahmane Cherif1
1Lab. des Mathématiques et Sciences Appliquées, University of Ghardaia, Bounoura, Algeria
2Laboratoire d’Informatique et des Mathematiques, University of Laghouat, Laghouat, Algeria
*Corresponding author. Email: habiba.benabderrahmane@lagh-univ.dz
Corresponding Author
Habiba Ben-Abderrahmane
Available Online 31 August 2024.
DOI
10.2991/978-94-6463-496-9_24How to use a DOI?
Keywords
Arabic Sign Language Recognition; LSTM; Video Processing; Keypoint Tracking
Abstract

Recognizing sign language is a vital task that assists in demystifying communication with deaf-mute persons. Many previous works tackle this problem by considering a static point of view, such as isolated single alphabet symbols or digit detection. This paper introduces an Arabic Sign Language (ArSL) recognition system using a deep learning technique. Our work raises the recognition level in two aspects: first, it deals with the dynamic nature of the problem and hence admits input from videos; and second, it experiments with a recent Arabic sign language dataset. Since only particular parts of the input convey the desired message, the proposed model pays attention to only the main regions in the input video and thus relies mainly on the use of keypoints of zones of interest tracked from video frame sequences. Regarding the sequence nature of the input data, the extracted keypoints are fed to an LSTM-based architecture specifically tailored to discover sentences from the input. Experiments on the ArabSign dataset reveal that our model succeeds in reproducing sentences existing in a sign video with an accuracy rate that surpasses 88.5%. The obtained results validate undeniably the effectiveness of the proposed model in recognizing a multitude of ArSL gestures.

Copyright
© 2024 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the International Conference on Emerging Intelligent Systems for Sustainable Development (ICEIS 2024)
Series
Advances in Intelligent Systems Research
Publication Date
31 August 2024
ISBN
978-94-6463-496-9
ISSN
1951-6851
DOI
10.2991/978-94-6463-496-9_24How to use a DOI?
Copyright
© 2024 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - Slimane Oulad-Naoui
AU  - Habiba Ben-Abderrahmane
AU  - Assia Chagha
AU  - Abderrahmane Cherif
PY  - 2024
DA  - 2024/08/31
TI  - An LSTM-based System for Dynamic Arabic Sign Language Recognition
BT  - Proceedings of the International Conference on Emerging Intelligent Systems for Sustainable Development (ICEIS 2024)
PB  - Atlantis Press
SP  - 313
EP  - 323
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-496-9_24
DO  - 10.2991/978-94-6463-496-9_24
ID  - Oulad-Naoui2024
ER  -