Video Recognition Techniques for Indian Sign Language in Healthcare Domain
Jash Gandhi, Parth Gandhi, Aayush Gosar, Sheetal Chaudhari · 2021 2nd International Conference for Emerging Technology (INCET) · 2021
Sign Language is an important mode of communication for the deaf-mute community. Sign language to text conversion models is currently in development and is getting popular day by day with the advancement of deep learning. With this paper, the plan is to create a system that can identify healthcare domain related basic words like doctor, blood, appointment, etc in real time using a webcam. In this paper, 3 different methods are compared for action recognition which gives an estimate of the output. Based on the comparative results, OpenPose + LSTM method is found to be the most effective. This will help in bridging the gap between the normal community and deaf-mute community.