VISION-BASED VIRTUAL MOUSE SYSTEM USING REAL-TIME HAND GESTURE RECOGNITION

Abstract

The Gesture Controlled Virtual Mouse is a cutting-edge computer vision-based technology that uses hand gestures to provide smooth, contactless mouse cursor control on a computer. The device uses sophisticated machine learning models to understand hand landmarks and dynamic finger movements while utilizing a simple webcam to collect real-time video. Key features include drag-and-drop (closed fist with movement), smooth scrolling (hand tilt detection), left-click (using the index finger pinch), rightclick (using the thumb and middle finger motion), and accurate cursor movement linked to hand position. The project, which is mostly written in Python, incorporates MediaPipe for hand landmark detection, PyAutoGUI for native mouse event simulation, and OpenCV for picture analysis. By removing physical contact, decreasing hardware wear, and encouraging hygienic, user-friendly interactions, this technology overcomes some of the main drawbacks of conventional input devices. It has a lot of potential for usage in medical applications (like sterile operating rooms), smart settings (like IoTintegrated houses), assistive technology for people with motor disabilities, and public kiosks where touchscreens could cause contamination. The technology shows viability for daily deployment by attaining real-time performance at 30+ frames per second on consumer-grade hardware, opening the door for more engaging human-computer interfaces.