Hand Gesture Controlled Presentation using OpenCV and MediaPipe

Abstract

In today's digital era, presentations play a crucial role in various domains, ranging from education to business. However, traditional manual presentation methods, reliant on input devices such as keyboards or clickers, have inherent limitations in terms of mobility, interactivity, and user experience. To address these limitations, gesture-controlled presentations have emerged as a promising solution, harnessing the power of computer vision techniques to interpret hand gestures and enable natural interaction with presentation content. This paper presents a comprehensive system for hand gesture-controlled presentations using OpenCV and MediaPipe libraries. OpenCV is employed to capture video input from a webcam, while MediaPipe is utilized for hand tracking and landmark extraction. By analyzing finger positions and movements, the system accurately recognizes predefined gestures. Presenters can seamlessly control the slides, hold a pointer, annotate the content, and engage with the audience in a more interactive manner. The responsiveness and real-time performance contribute to an enhanced presentation experience.