Scholarlink

<p>The rising demand for contactless and effortless human–computer interaction has driven the shift from traditional physical mouse devices toward gesture-based control. This study introduces a machine-learning-powered virtual mouse that enables complete laptop operation using real-time hand gestures captured through a standard webcam—eliminating the need for hardware peripherals. The system integrates MediaPipe-based hand tracking with OpenCV and a machine learning model to interpret fingertip positions and convert gestures into operations such as left click, right click, drag-and-drop, scrolling, and brightness/volume adjustment. The approach is designed for real-time performance, intuitive usability, and accessibility for users including individuals unable to operate a physical mouse. Experimental evaluation demonstrates an operational accuracy of 97%, outperforming existing solutions and confirming its suitability for daily computing scenarios and touchless interaction environments. Overall, the proposed work offers a practical, affordable, and scalable alternative to conventional pointing devices and contributes a distinctive enhancement to gesture-based human–computer interaction.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Beyond the Click: A Machine Learning Powered Gesture Recognition for Enhanced Laptop Interaction

  • 期刊论文

  • 发表日期：

    2026年6月3日

查看全文
  • Ratnesh Litoriya,
  • Kailash Chandra Bandhu,
  • Arpit Deo,
  • Prashant Panse,
  • Richa Trivedi,
  • Rashi Bagla

摘要

The rising demand for contactless and effortless human–computer interaction has driven the shift from traditional physical mouse devices toward gesture-based control. This study introduces a machine-learning-powered virtual mouse that enables complete laptop operation using real-time hand gestures captured through a standard webcam—eliminating the need for hardware peripherals. The system integrates MediaPipe-based hand tracking with OpenCV and a machine learning model to interpret fingertip positions and convert gestures into operations such as left click, right click, drag-and-drop, scrolling, and brightness/volume adjustment. The approach is designed for real-time performance, intuitive usability, and accessibility for users including individuals unable to operate a physical mouse. Experimental evaluation demonstrates an operational accuracy of 97%, outperforming existing solutions and confirming its suitability for daily computing scenarios and touchless interaction environments. Overall, the proposed work offers a practical, affordable, and scalable alternative to conventional pointing devices and contributes a distinctive enhancement to gesture-based human–computer interaction.