CASL-W60: A word-level dataset for central African sign language recognitionKaggle

Sign language is a non-verbal discourse system used by people who are hard of hearing. It also carries cultural context and regional constructs, enabling meaningful communication and often preserving unique traditions. In the Central African region, local sign languages have distinct linguistic cons...

Full description

Saved in:
Bibliographic Details
Main Authors: Mwaka Lucky, Njayou Youssouf, Hasan Mahmud, Md Kamrul Hasan
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Data in Brief
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352340925005177
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849249370855178240
author Mwaka Lucky
Njayou Youssouf
Hasan Mahmud
Md Kamrul Hasan
author_facet Mwaka Lucky
Njayou Youssouf
Hasan Mahmud
Md Kamrul Hasan
author_sort Mwaka Lucky
collection DOAJ
description Sign language is a non-verbal discourse system used by people who are hard of hearing. It also carries cultural context and regional constructs, enabling meaningful communication and often preserving unique traditions. In the Central African region, local sign languages have distinct linguistic constructs but remain underrepresented in the literature, creating a significant gap in regional word-level datasets for machine learning practitioners. In this research, we present a dataset (CASL-W60) comprising 60 word-level Central African sign language (CASL), collected from 19 volunteers. Each word contains 10–12 video samples per signer, captured following standard African sign language video references. The dataset comprises MP4 video files that are systematically organized and made available through an online repository. We demonstrate its applicability through word-level classification of the 60 sign words. This dataset serves as a valuable resource for developing various applications, including sign language translation, sentence recognition or generation from word-level signs, and sign gloss detection.
format Article
id doaj-art-1101a811a36c4ab2a9a28b2e2dcdbfd6
institution Kabale University
issn 2352-3409
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series Data in Brief
spelling doaj-art-1101a811a36c4ab2a9a28b2e2dcdbfd62025-08-20T03:57:36ZengElsevierData in Brief2352-34092025-08-016111179010.1016/j.dib.2025.111790CASL-W60: A word-level dataset for central African sign language recognitionKaggleMwaka Lucky0Njayou Youssouf1Hasan Mahmud2Md Kamrul Hasan3Systems and Software Lab (SSL), Department of Computer Science Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur 1704, Dhaka, BangladeshSystems and Software Lab (SSL), Department of Computer Science Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur 1704, Dhaka, BangladeshSystems and Software Lab (SSL), Department of Computer Science Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur 1704, Dhaka, BangladeshCorresponding author.; Systems and Software Lab (SSL), Department of Computer Science Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur 1704, Dhaka, BangladeshSign language is a non-verbal discourse system used by people who are hard of hearing. It also carries cultural context and regional constructs, enabling meaningful communication and often preserving unique traditions. In the Central African region, local sign languages have distinct linguistic constructs but remain underrepresented in the literature, creating a significant gap in regional word-level datasets for machine learning practitioners. In this research, we present a dataset (CASL-W60) comprising 60 word-level Central African sign language (CASL), collected from 19 volunteers. Each word contains 10–12 video samples per signer, captured following standard African sign language video references. The dataset comprises MP4 video files that are systematically organized and made available through an online repository. We demonstrate its applicability through word-level classification of the 60 sign words. This dataset serves as a valuable resource for developing various applications, including sign language translation, sentence recognition or generation from word-level signs, and sign gloss detection.http://www.sciencedirect.com/science/article/pii/S2352340925005177MediaPipe hand landmarksRegional sign language recognitionVideo framesMachine learning
spellingShingle Mwaka Lucky
Njayou Youssouf
Hasan Mahmud
Md Kamrul Hasan
CASL-W60: A word-level dataset for central African sign language recognitionKaggle
Data in Brief
MediaPipe hand landmarks
Regional sign language recognition
Video frames
Machine learning
title CASL-W60: A word-level dataset for central African sign language recognitionKaggle
title_full CASL-W60: A word-level dataset for central African sign language recognitionKaggle
title_fullStr CASL-W60: A word-level dataset for central African sign language recognitionKaggle
title_full_unstemmed CASL-W60: A word-level dataset for central African sign language recognitionKaggle
title_short CASL-W60: A word-level dataset for central African sign language recognitionKaggle
title_sort casl w60 a word level dataset for central african sign language recognitionkaggle
topic MediaPipe hand landmarks
Regional sign language recognition
Video frames
Machine learning
url http://www.sciencedirect.com/science/article/pii/S2352340925005177
work_keys_str_mv AT mwakalucky caslw60awordleveldatasetforcentralafricansignlanguagerecognitionkaggle
AT njayouyoussouf caslw60awordleveldatasetforcentralafricansignlanguagerecognitionkaggle
AT hasanmahmud caslw60awordleveldatasetforcentralafricansignlanguagerecognitionkaggle
AT mdkamrulhasan caslw60awordleveldatasetforcentralafricansignlanguagerecognitionkaggle