Two stream GRU model with ELU activation function for sign language recognition

Pose Estimation features have been successfully used in human activity recognition including sign language recognition. One of the key challenges in sign language recognition is handling signer-independent modes and hand dominance of signer. This paper proposes the use of the Gated Recurrent Unit (G...

Full description

Saved in:
Bibliographic Details
Main Authors: Kasian Myagila, Devotha Godfrey Nyambo, Mussa Ally Dida
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Intelligent Systems with Applications
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2667305325000390
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728779003363328
author Kasian Myagila
Devotha Godfrey Nyambo
Mussa Ally Dida
author_facet Kasian Myagila
Devotha Godfrey Nyambo
Mussa Ally Dida
author_sort Kasian Myagila
collection DOAJ
description Pose Estimation features have been successfully used in human activity recognition including sign language recognition. One of the key challenges in sign language recognition is handling signer-independent modes and hand dominance of signer. This paper proposes the use of the Gated Recurrent Unit (GRU) with the ELU activation function to improve computation efficiency and to enhance model learning efficiency. In addition, the paper proposes two stream model architecture to address the challenge of left and right-hand dominance. The study developed model using a Tanzania Sign language datasets collected using mobile devices and extracted pose estimation feature using MediaPipe holistic framework. According to the results, the proposed model not only achieves an impressive overall accuracy of 95%, but also trains more efficiently than comparable algorithms. Particularly in the signer-independent mode, the two-stream approach led to substantial improvements, achieving a maximum accuracy of 92% and a minimum accuracy of 70% with significant increase on the left handed signer accuracy by 37%. The results highlight the effectiveness of the two-stream approach in overcoming challenges related to left and right-hand dominance, which often arise from signer-specific hand dominance. Additionally, the results indicate that, the proposed model can have a positive impact on limited computational resources while also enhancing the model’s overall performance.
format Article
id doaj-art-abaa178594cf44a7bbae718977788036
institution DOAJ
issn 2667-3053
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Intelligent Systems with Applications
spelling doaj-art-abaa178594cf44a7bbae7189777880362025-08-20T03:09:27ZengElsevierIntelligent Systems with Applications2667-30532025-06-012620051310.1016/j.iswa.2025.200513Two stream GRU model with ELU activation function for sign language recognitionKasian Myagila0Devotha Godfrey Nyambo1Mussa Ally Dida2School of Computation and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, P.O. Box 447, Arusha, Tanzania; Department of Computing Science Studies, Mzumbe University, P.O. Box 87, Morogoro, Tanzania; Corresponding author at: Department of Computing Science Studies, Mzumbe University, P.O. Box 87, Morogoro, Tanzania.School of Computation and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, P.O. Box 447, Arusha, TanzaniaSchool of Computation and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, P.O. Box 447, Arusha, TanzaniaPose Estimation features have been successfully used in human activity recognition including sign language recognition. One of the key challenges in sign language recognition is handling signer-independent modes and hand dominance of signer. This paper proposes the use of the Gated Recurrent Unit (GRU) with the ELU activation function to improve computation efficiency and to enhance model learning efficiency. In addition, the paper proposes two stream model architecture to address the challenge of left and right-hand dominance. The study developed model using a Tanzania Sign language datasets collected using mobile devices and extracted pose estimation feature using MediaPipe holistic framework. According to the results, the proposed model not only achieves an impressive overall accuracy of 95%, but also trains more efficiently than comparable algorithms. Particularly in the signer-independent mode, the two-stream approach led to substantial improvements, achieving a maximum accuracy of 92% and a minimum accuracy of 70% with significant increase on the left handed signer accuracy by 37%. The results highlight the effectiveness of the two-stream approach in overcoming challenges related to left and right-hand dominance, which often arise from signer-specific hand dominance. Additionally, the results indicate that, the proposed model can have a positive impact on limited computational resources while also enhancing the model’s overall performance.http://www.sciencedirect.com/science/article/pii/S2667305325000390Sign language recognitionGRUELU functionPose estimationSigner independent
spellingShingle Kasian Myagila
Devotha Godfrey Nyambo
Mussa Ally Dida
Two stream GRU model with ELU activation function for sign language recognition
Intelligent Systems with Applications
Sign language recognition
GRU
ELU function
Pose estimation
Signer independent
title Two stream GRU model with ELU activation function for sign language recognition
title_full Two stream GRU model with ELU activation function for sign language recognition
title_fullStr Two stream GRU model with ELU activation function for sign language recognition
title_full_unstemmed Two stream GRU model with ELU activation function for sign language recognition
title_short Two stream GRU model with ELU activation function for sign language recognition
title_sort two stream gru model with elu activation function for sign language recognition
topic Sign language recognition
GRU
ELU function
Pose estimation
Signer independent
url http://www.sciencedirect.com/science/article/pii/S2667305325000390
work_keys_str_mv AT kasianmyagila twostreamgrumodelwitheluactivationfunctionforsignlanguagerecognition
AT devothagodfreynyambo twostreamgrumodelwitheluactivationfunctionforsignlanguagerecognition
AT mussaallydida twostreamgrumodelwitheluactivationfunctionforsignlanguagerecognition