Published Paper Details:

The Sign Language Detection system is developed for detecting the alphabets of American Sign Language (ASL). Conventionally, sign languages comprise of finger spelling. For detecting the signs, the Regions of Interest (ROI) are related and tracked employing the Landmarks feature of Media Pipe. Then by using Open CV, we capture the landmarks of the hands, and the key points of landmarks are stored in a NumPy array. Then we can train the model on it by using TensorFlow, Keras, and CNN. Finally, the model can be tested in real-time by ingesting live feed incoming from the webcam. Realtime Sign Detection is one of the essential applications to be used by deaf and dumb people as it helps them to communicate with others effectively. Historically, various methodologies for detecting signs were employed by the Machine Learning Algorithm, by training it on the image data . However, now we are employing Deep Learning Models to simplify and speed up the process of real-time sign detection and recognition and produce equal or more accuracy using smaller amounts of data.