Deep Learning-Based Sign Language Recognition System

Abstract

The area of hand gesture recognition has many interesting applications. An important application of hand gesture recognition is the translation of sign language, which is a complex structured form of hand movements. A rigorous experimental evaluation of computer vision-based methods for sign language recognition is the goal of this research article. To understand sign language recognition, the current study focuses on mapping non-segmented video streams to glosses. Long Short-Term Memory (LSTM) is one of the Recurrent Neural Network (RNN) layers that make up the proposed machine learning model. Current Deep learning frameworks, such as Google Tensor Flow and Keras API, are used to implement the model.