Review on Deep Learning-Based Dynamic Indian Sign Language Recognition System

Abstract

An alternative form of communication for those who are hard of hearing or speech is sign language (SL). SL differs from nation to nation. Indian Sign Language (ISL) is a substitute form of communication used by individuals in India. There are about 63 million deaf and hard of hearing persons living in India. But there are no accessible or reasonably priced human interpreters for ISL identification. Consequently, for many years, researchers have been working on creating a system that can recognize sign language. Compared to other SL, the construction of an Indian Sign Language Recognition System (SLRS) is exceedingly challenging. The field of Indian sign language recognition systems has not seen many developments. Consequently, this discipline has very few systematic literature reviews. We reviewed the most recent (post-2017) literature on dynamic Indian sign language recognition using deep learning technique on various parameters in an attempt to tackle this problem. The objective of this work is to present a comprehensive analysis of various deep learning methods used in dynamic Indian sign language recognition.