Empowering Communication: Harnessing CNN and Mediapipe for Sign Language Interpretation

Sign language serves as a crucial communication medium for the Deaf and Hard-of-Hearing community. However, the lack of automated systems for sign language recognition can limit interaction with people who are unfamiliar with it. This study proposes a real-time sign language detection system leveraging OpenCV for image preprocessing, MediaPipe for hand tracking, and Convolutional Neural Networks (CNN) for gesture classification. The system captures video feed, isolates hand regions, and identifies gestures that correspond to alphabets in American Sign Language (ASL). Through this model, it has achieved a promising classification accuracy, thereby showing significant promise for real-world applications. This work aims to bridge the communication gap by facilitating seamless interaction between hearing impaired individuals and the broader community, enhancing inclusivity and accessibility.