Gesture Controlled Virtual Mouse: A Review and Camera-Based System Design

A conventional mouse is the standard tool for moving a pointer, but hand gestures offer another way to interact with a computer. The availability of low-cost webcams and efficient hand-tracking models has made this approach more practical. This paper reviews major methods for gesture-based mouse control, including wearable sensors, depth cameras, colour markers, skin segmentation, neural-network detectors and landmark tracking, and discusses their advantages and drawbacks. It also presents the analysis and specification of a webcam-based virtual mouse, covering use cases, data flow, processing stages, system states and feasibility. The proposed design uses a hand-landmark model to estimate 21 points in each frame. Rules based on finger positions identify gestures and translate them into cursor movement, clicking, scrolling and drag-and-drop operations. Key challenges include changing illumination, pointer jitter, unintended commands and arm fatigue; possible improvements are also discussed.