用于虚拟现实/增强现实人机交互的实时手势识别方法的研究  

Research on real-time gesture recognition for human-computer interaction in virtual reality/augmented reality

作　　者：柯健[1] 张量[1] Ke Jian;Zhang Liang(School of Computer Engineering,Suzhou Vocational University,Suzhou 215104)

机构地区：[1]苏州市职业大学计算机工程学院,苏州215104

出　　处：《现代计算机》2023年第1期59-62,共4页Modern Computer

摘　　要：虚拟现实/增强现实(简称VR/AR)中的人机交互方式是一个重要的研究方向,将手势与VR/AR中的操作控制相结合,提出一种利用手势实现VR/AR操作的人机交互方式。在实验中使用MediaPipe框架采集实时手势图像,提取手部关键点,设计手势关键点算法进行实时手势识别,结果显示该方法具有良好的性能,为VR/AR提供了一种更为自然的人机交互方式。Human-computer interaction in virtual reality/augmented reality is an important research direction. This paper combines gestures with operation control in VR/AR, and proposes a human-computer interaction method that uses gestures to realize VR/AR operation. In the experiment, the MediaPipe framework was used to collect real-time gesture images, extract hand key points, and design a gesture key point algorithm for real-time gesture recognition. The results show that the method has good performance and provides a more natural human-computer interaction mode for VR/AR.

关 键 词：手势识别 人机交互 手部关键点检测 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]