随着信息技术的发展以及信息技术在教育教学中的深入运用，课堂上学生和教师除了课堂教学内容外还多了很多需要注意的东西，比如PPT、电子白板等，教师能否关注到每一位学生至关重要，因此对于教师走位的研究必不可少。但至今在教育教学中关于教师课堂走位的专门研究还很少。同时，教师走位的研究更多是通过课堂观察和访谈，较为费时费力。使用计算机视觉进行课堂教学视频教师走位分析的研究成果还不够丰富。
       本文的研究内容是基于人脸识别技术，结合YOLOv5和ArcFace算法研究了课堂教学视频中教师走位分析系统的搭建。本文通过使用CelebA数据集重新训练人脸检测模型YOLOv5s网络提取出人脸，使用CASIA-webface和LFW数据集训练和验证的ArcFace模型进行人脸识别，经过需求分析、设计、实现和测试，完成了人脸检测识别的实现、图像化界面设计、教师走位的监测记录，基本实现了基于人脸识别的课堂教学视频教师走位分析系统的搭建。基于人脸识别的课堂教学视频教师走位分析系统通过对课堂教学视频进行分析，实现了教师人脸识别与图像定位、基于学生座位/教室长宽尺寸/黑板尺寸的教师粗略走位分析，为教师的课后反思、量化评价及职业规划发展提供数据支持，丰富了课堂教学视频和教师课堂评价研究理论体系。

      As information technology becomes increasingly integrated into education, students and teachers must also be mindful of tools beyond traditional classroom content, such as PPT and electronic whiteboards. It is crucial for teachers to focus on every student, therefore, research on teacher's position and walking route in the classroom is essential, but there is still little specialized research on this topic in education. Additionally, current research on teacher's position and walking route mostly relies on classroom observations and interviews, which can be time-consuming and laborious. The research in using computer vision to analyze teacher's position and walking route in classroom teaching videos is not yet well-developed.
      The research content of this paper is based on face recognition technology, combined with YOLOv5 and ArcFace algorithm to study the construction of teacher position and walking route analysis system in classroom teaching videos. In this paper, the open source face data set CelebA retrained YOLOv5s network is used as the face detection model to extract the face region image, and ArcFace model trained and verified by CASIA-webface and LFW data sets is used for face recognition. After system requirements analysis, design, implementation and testing, completed the realization of face detection and recognition, graphical interface design, teacher position monitoring record, basically realized the construction of teacher position and walking route analysis system in classroom teaching videos based on face recognition. Through the analysis of classroom teaching videos, the teacher position and walking route analysis system based on face recognition realizes the teacher's face recognition and image positioning, and the teacher position and walking route analysis based on student seat/classroom length and width/blackboard size, which provides reference for teachers' after-class reflection, quantitative evaluation and career planning development. Enriched the classroom teaching video and teacher classroom evaluation research theory system.