Sign Language Detecting System Using Tensorflow Object Detection API

Communication is the only way we can communicate or express ourselves, but for those with disabilities like deafness and neuropathy face challenges in communicating with others. To enhance communication between deaf individuals and regular people, we developed a sign language detection system that employs TensorFlow's object detection capabilities to enable real-time translation of sign Language gestures. Primarily hand gestures are obtained from the deaf and hard-of-hearing people and then compared with the data set. The data set formerly consist of n number of gesture gesticulations which are already labeled and stored with the help of Single Shot MultiboxDetector (SSD) concept. Once the features of an incoming image are paralleled with features of formerly labeled images in a data set, it indicates recognition of gesture. Eventually, the gesture is translated into the message. The accessibility of communication was improved by this sign language detecting system. This assistive technology can be utilized in future applications due to its precise and efficient real time sign language detection. It provides a real-time means of converting hand movements into text