AIR-CANVAS: A Computer Vision Based Gesture Drawing System
Kiran Ande · International Journal for Research in Applied Science and Engineering Technology · 2025
Abstract: The Air Canvas Project is an innovative computer vision application developed using OpenCV and Python, which enables users to draw in the air using simple hand gestures tracked via a webcam. By utilizing a coloured marker—typically placed at the tip of a finger—the system captures the motion of the marker in real time and renders corresponding strokes on a virtual canvas. The project leverages key techniques in computer vision, including colour detection in HSV colour space, contour tracking, and morphological operations like erosion and dilation to enhance accuracy and reduce noise. The system continuously tracks the position of the coloured marker, stores the coordinates, and draws the path traced by the user across successive video frames. This approach eliminates the need for traditional input devices, offering a touchless, interactive experience. Designed with accessibility and simplicity in mind, the Air Canvas can serve as both an educational tool for learning computer vision and a creative platform for digital expression. The project demonstrates the potential of combining real-time image processing and gesture recognition to create intuitive, user-friendly applications.