Scholarlink

<p>In the field of basic education, monitoring and evaluating students’ attention can assist teachers in understanding their engagement levels during the teaching process. Existing studies have employed machine learning, artificial intelligence, and other techniques to evaluate classroom attention. However, these methods often neglect the multidimensional characteristics of students’ attention. Therefore, this study aims to develop a method for attention monitoring and evaluation that accurately reflects real-time changes in students’ attention during lessons. Key features of both teachers’ and students’ facial expressions were analyzed. Students’ attention was depicted through a rhythm spectrum, and their overall attention was evaluated by combining face detection with head pose estimation. The results indicate that students’ attention presents obvious phased changes, with the golden periods (from the 5th to the 20th minute) and low periods (from the 20th to the 25th minute and the last 5 minutes). This study monitors and evaluates students’ attention levels during a 50-minute class, providing a reference for subsequent students’ intervention. Furthermore, this study enables dynamic teaching adjustments by offering real - time insights into students’ attention states, thereby enhancing teaching quality.</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

A case study of monitoring and evaluation method of students’ attention using face detection and head pose Estimation in a classroom

  • 期刊论文

  • 发表日期：

    2025年10月6日

查看全文
  • Zhongxiang Feng,
  • Siqing Zhang,
  • Li Dai,
  • Zeyang Cheng,
  • Zhipeng Huang

摘要

In the field of basic education, monitoring and evaluating students’ attention can assist teachers in understanding their engagement levels during the teaching process. Existing studies have employed machine learning, artificial intelligence, and other techniques to evaluate classroom attention. However, these methods often neglect the multidimensional characteristics of students’ attention. Therefore, this study aims to develop a method for attention monitoring and evaluation that accurately reflects real-time changes in students’ attention during lessons. Key features of both teachers’ and students’ facial expressions were analyzed. Students’ attention was depicted through a rhythm spectrum, and their overall attention was evaluated by combining face detection with head pose estimation. The results indicate that students’ attention presents obvious phased changes, with the golden periods (from the 5th to the 20th minute) and low periods (from the 20th to the 25th minute and the last 5 minutes). This study monitors and evaluates students’ attention levels during a 50-minute class, providing a reference for subsequent students’ intervention. Furthermore, this study enables dynamic teaching adjustments by offering real - time insights into students’ attention states, thereby enhancing teaching quality.