基于OpenCV与MediaPipe的面部动作控制鼠标操作技术  

Research on Facial Action Control Mouse Operation Based on OpenCV and MediaPipe

作　　者：王骏祥 马健为 宋笛秋 朱思哲 WANG Junxiang;MA Jianwei;SONG Diqiu;ZHU Sizhe(College of Software,Nankai University,Tianjin 300457,China)

机构地区：[1]南开大学软件学院,天津300457

出　　处：《计算技术与自动化》2024年第2期177-181,共5页Computing Technology and Automation

摘　　要：为丰富人机交互模式,基于OpenCV和Mediapipe在面部识别中的应用,在个人电脑上实现通过面部动作进行常用的鼠标控制。具体的方法是通过捕捉摄像头传入视频帧中用户脸部的关键点,并根据关键点的位置变化来执行对应鼠标操作。实际测试结果表明:该程序仅通过简单的面部动作就可以较好地实现诸如移动光标、单击鼠标左右键以及拖动滚动条等常用鼠标操作,且识别准确率高,是一种针对特殊用户和特定环境条件下替代鼠标操作的低成本实用技术。In order to enrich the human-computer interaction methods,this work realizes common mouse control through facial movements on PC based on the application of OpenCV and MediaPipe in facial recognition.The specific method is to capture the key points of user’s face in the video frame transmitted by the camera,and perform corresponding operations according to the position changes of the key points.The actual test results show that this program can effectively achieve common mouse operations such as moving the cursor,clicking the left and right mouse buttons,and dragging the scroll bar through simple facial movements,and has high recognition accuracy.It is a low-cost practical technology that can replace mouse operations for special users and specific environmental conditions.

关 键 词：MediaPipe OPENCV 面部识别 鼠标操作 

分 类 号：TP317.1[自动化与计算机技术—计算机软件与理论]