Arabic sign language recognition with 3D convolutional neural networks

Sign Language recognition is very important for communication purposes between Hearing Impaired (HI) people and hearing ones. Arabic Sign Language Recognition field became widespread because of its difficult nature and numerous details. Most researchers employed different input sensors, features extractors, and classifiers on static and dynamic data. These different ways were customized and employed in our previous work in the Arabic Sign Language Recognition field. In this paper, features extractor with deep behavior was used to deal with the minor details of Arabic Sign Language. 3D Convolutional Neural Network (CNN) was used to recognize 25 gestures from Arabic sign language dictionary. The recognition system was fed with data from depth maps. The system achieved 98% accuracy for observed data and 85% average accuracy for new data. The results could be improved as more data from more different signers are included.