Sign Language Translation using Tensor Flow Model Zoo

Abstract

Communication is very essential for all humans because it allows us to share or express our sentiments, emotions, and other thoughts. Humans communicate with one another via natural language (for example words), body language (hand gestures, facial motions, and so on) or writing etc. People who do not have any impairment can easily converse with each other in natural language. However people with impairments, such as deafness or blindness, suffer a communication hurdle. They converse with common people by adopting sign languages, which are difficult for normal people to grasp or understand. People with hearing and speech impairments often have a very difficult time conversing with other people without any translator or interpreter. As a matter of fact, this research is being undertaken that transforms sign language into text that is easily understood by the ordinary individual. This system will identify numbers, alphabets, and hand gestures as well. Our primary purpose is to remove the obstacles that exist between the deaf, dumb, and the rest of people.