AI based Smart Conferencing System using Gesture Recognition

The paper proposes the use of computer vision to translate sign language into text and audio, which can aid in communication for individuals with hearing loss. The system consists of four modules: picture capture, pre-processing, classification, and prediction. The segmentation of sign motions is achieved through image processing using Media Pipe in conjunction with the OpenCV Python package. To ensure high accuracy in predicting sign language, captured motions are scaled, converted to greyscale, and filtered for noise. The SVM classifier is utilized for classification and prediction. Additionally, natural language processing is employed to create meaningful sentences from the recognized sign gestures. Overall, this initiative has the potential to be a useful tool for those who have hearing loss and might greatly improve their capacity for interpersonal communication.