Real-Time Sign Language Recognition
Rohan Gupta, Chirag Pandit, Krishnam Gupta, Prabhjot Singh
Abstract
Communication refers to exchanging information between two people to perform different tasks or develop bonding for personal or professional use. Effective communication provides a lot of advantages in terms of business meetings and gatherings. It is quite easy for normal people without these disability to communicate with each other but there is a major problem of communication for deaf people all over the world. Sign language is a major source of communication for deaf people, but the problem arises for normal people to communicate with them, so sign language acts as a source of communication for them. A model for the recognition of various gestures has been designed for this problem. LSTM (long short-term memory) model has been proposed for sign language recognition. The model is designed in a way that it can take input from the user through the webcam which has been made with the help of datasets fed with the help of OpenCV which has been trained to recognize the datasets through labels set on them. The output of the software is that the user can thereby see the meaning of the symbol or signs that has been performed in front of the webcam. This model can be quite effective in terms of exchanging information between deaf and mute people and normal people.