A study of the Sign language identifying using the OpenCV, MediaPipe, and Scikit-learn modules

Sign language is manual communication commonly used by people who are hard of speaking and hearing. These languages use the visual-manual modality to convey meaning, instead of spoken words. Sign languages are expressed through manual articulation using hand gestures, facial expressions, and body language to describe the intended message as well as some non-manual markers. This paper proposes a novel approach to interpreting sign language using the camera of a phone or a laptop breaking the communication barrier between a mute person and a person who does not know sign language. In this approach, the model has been trained to identify some signs using the OpenCV, MediaPipe, and Scikit-learn modules. The hand landmarks from the image data set were extracted using the media pipe module to train the model. The model can identify signs and once the sign has been identified it can play the corresponding sign in the form of audio.