Indian Sign Language Recognising Web Application

Abstract

Recognition of sign language is a problem that has been studied for many years. In the related
discipline of American Sign Language (ASL), a lot of study has been done, but the same
cannot be said for ISL. The lack of standard datasets, occluded features, and linguistic
heterogeneity with locale have been the key roadblocks, resulting in minimal ISL study.
However, we are still a long way from finding a comprehensive solution in our culture. Using
deep learning, we are developing a real-time sign detecting model. Our system detects a
person's sign language, recognizes the person's sign or position, and converts it into text using
this model. We offer a module in which videos are displayed. The signs are detected and
transformed into text that everyone can understand using pose estimation as the input and the
Convolutional Neural Network (CNN) method.