Real-Time Sign Language Recognition and Translation

CH, Bhupati, Garre Gayatri, Vudata Harika, Vemula Thanuja

doi:10.7492/0tfdw884

Authors

CH, Bhupati, Garre Gayatri, Vudata Harika, Vemula Thanuja Author

DOI:

https://doi.org/10.7492/0tfdw884

Abstract

Communication is the primary way to interact with each other.It is difficult for people suffering with hearing impaired individuals for communication.To help close this gap, this paper introduces a model which works in real time that makes communication and translation between signers and non- signers smooth and natural. Instead of relying on expensive gloves or sensors, this system simply uses a camera and a deep learning model built with TensorFlow and OpenCV. The model, based on SSD MobileNet V2, can quickly recognize different hand gestures

both still and moving — from a live video feed.Every frame of the video is processed in real time, and the identified gesture is immediately converted to text and voiced through a Text-to- Speech (TTS) module.