Sign Language Recognition Using Convolutional Neural Networks

Abstract

Abstract-Sign language is a lingua among the speech and the hearing impaired community. It is hard for most people who are not familiar with sign language to communicate without an interpreter. Sign language recognition appertains to track and recognize the meaningful emotion of human made with fingers, hands, head, arms, face etc. The technique that has been proposed in this work, transcribes the gestures from a sign language to a spoken language which is easily understood by the hearing. The gestures that have been translated include alphabets, words from static images. This becomes more important for the people who completely rely on the gestural sign language for communication tries to communicate with a person who does not understand the sign language. We aim at representing features which will be learned by a technique known as convolutional neural networks (CNN), contains four types of layers: convolution layers, pooling/subsampling layers, non-linear layers, and fully connected layers. The new representation is expected to capture various image features and complex non-linear feature interactions. A softmax layer will be used to recognize signs.