A Comprehensive Review on Sign Language Recognition Using Machine Learning
Janpreet Singh, Dalwinder Singh
Abstract
The people of the dumb and deaf community adopt sign language to talk with one another. Sign language is an effective method for the interaction between speech and hearing-impaired individuals and a normal person without any requirement of an interpreter. Every nation has its own sign language, which has been used by the impaired residents of that particular country to communicate either with normal individuals or with a computer system in real-time. The recognition of sign language becomes an emerging research domain in computer vision. The researchers find several difficulties to recognize them correctly and adequately as the sign instance varies with appearance as well as motion. This paper explained all the challenges faced by the researchers while developing or designing a real-time system for sign language recognition. The main intent of this paper is to review the numerous machine learning methodologies that are utilized by various researchers to present a real- time system for the recognition of sign language. The performances of the existing system are also compared with each other to identify which machine learning methodology is generating accurate as well as an effective result. Based on this review paper, further future research work can be conducted as this paper also presented the directions of present future research.