Air canvas: a real-time hand gesture-based drawing interface using computer vision

This research presents Air Canvas, an innovative real-time, gesture-based virtual drawing system that enables users to draw in the air using hand movements, eliminating the need for physical contact with input devices. The system uses a standard webcam and applies computer vision techniques to detect and track a user’s fingertip, typically marked with a colored object like a marker or tape. Utilizing color segmentation in the HSV (Hue, Saturation, Value) color space, along with contour detection and trajectory mapping, Air Canvas accurately captures hand gestures and translates them into continuous strokes on a digital canvas. Developed using Python and OpenCV, the system ensures efficient real-time processing and cross-platform functionality. The application includes modules for video input, preprocessing, color detection, and drawing logic. Its non-contact design makes it especially useful in hygiene-sensitive environments such as hospitals, or for users with limited mobility who may find traditional input methods difficult to use. Performance tests showed that the system consistently operates at 25–30 FPS, delivering smooth and responsive drawing with minimal latency. Despite challenges in poor lighting or similar-colored backgrounds, the system proved reliable. The study also discusses future enhancements, such as integrating deep learning for hand tracking and gesture-based control commands, and expanding to mobile or web-based platforms.