Arabic Alphabets Sign Language Dataset (ArASL)

Published: 5 November 2018| Version 1 | DOI: 10.17632/y7pckrw6z2.1
Contributors:
Ghazanfar Latif, Jaafar Alghazo, Nazeeruddin Mohammad, Roaa AlKhalaf, Rawan AlKhalaf

Description

A new dataset consists of 54,049 images of ArSL alphabets performed by more than 40 people for 32 standard Arabic signs and alphabets. The number of images per class differs from one class to another. Sample image of all Arabic Language Signs is also attached. The CSV file contains the Label of each corresponding Arabic Sign Language Image based on the image file name.

Files

Institutions

Universiti Malaysia Sarawak, Prince Mohammad Bin Fahd University

Categories

Image Processing, Image Classification, Sign Language, Arabic Language, Recognition

Licence