Mime Recognition for Indian Sign Language

Abstract

Signs are part of non-verbal conveyance. Deaf and mute people primarily use sign language to interact with one another. Communicating with them has always been a significant challenge for the general public, as gestures are difficult to understand. Therefore, the important idea is to assist the com- munication absence among the public and the hearing impaired. Various sign language systems have been developed, but they are neither flexible nor cost-effective. Therefore, this project proposes an effective and user-friendly sign language recognition interface that makes it easy for hearing-impaired people to communicate with the general public using Tensorflow, Keras, and CNN (convolutional neural networks) for gesture recognition.