Hand Gesture Recognition and Translation for International Sign Language Communication using Convolutional Neural Networks
S Sivamohan, S. Anslam Sibi, T Divakar, S. Jagan
Abstract
A fundamental and natural form of communication, hand gestures are especially important for those who struggle with hearing loss. This Research provides a novel real-time technique for understanding hand motions used in sign language, specifically for International Sign Language (ISL), by using Convolutional Neural Networks (CNNs). This ground-breaking method aims to translate hand motions seen on a live video feed, making communication easier for those who use sign language. It’s amazing how well the system recognizes ISL movements and converts them into audible speech, removing a major barrier to communication between those who use sign language and others who don’t. This technology innovation promotes inclusivity by facilitating effortless communication for people with all linguistic and auditory skills. Extensive testing conducted under various settings highlights the system’s resilience, exhibiting a remarkable accuracy rate of 97.85%. This accomplishment confirms the system’s dependability in practical applications by highlighting its feasibility despite changing background intricacies and illumination subtleties.