Gesture-Based Air Typing: A Machine Learning Approach for Accessible Text Input

Abstract

The "Gesture-Based Air Typing: A Machine Learning Approach for Accessible Text Input" project represents a groundbreaking approach to hands-free typing, designed to revolutionize human-computer interaction by leveraging advanced hand gesture recognition technologies. This innovative system employs MediaPipe for precise hand tracking and EasyOCR for accurate text recognition, enabling users to seamlessly input text through freehand gestures captured in realtime using a simple webcam. The project's emphasis on cost-effectiveness and accessibility ensures that it caters to a broad audience, including individuals with physical impairments and those seeking futuristic, touch-free typing solutions. By eliminating the reliance on physical keyboards, this system opens new possibilities for inclusive design and interaction in digital environments.