Authors: Vishnu Koudgave, Pratik Londhe, Anishka Ahuja, Prasanna Kharbas, Prof. Jyoti Raghatwan
Abstract: With the growing demand for touchless and intelligent computing systems, hand gesture recognition has emerged as an innovative approach for natural human-computer interaction. This paper presents an AI-based Virtual Mouse system that enables real-time cursor control using hand gestures captured through a webcam. The proposed system utilizes MediaPipe for detecting 21 hand landmarks and OpenCV for real-time video processing, while PyAutoGUI is used to perform mouse operations such as cursor movement, clicking, scrolling, and dragging. The system provides smooth, accurate, and low-latency interaction without requiring additional hardware, making it a cost-effective and user-friendly solution. The proposed model enhances touchless human-computer interaction and has potential applications in smart environments, virtual reality systems, gaming, and assistive technologies.
DOI: https://doi.org/10.5281/zenodo.20444460
