Scholarlink

Effective classroom management relies on accurate monitoring of student behaviors, yet traditional manual observation is subjective and labor-intensive. This study leverages YOLOv8, a state-of-the-art object detection framework, to automatically detect behaviors such as “Focused”, “Raising Hand”, “Distracted”, “Sleep”, and “Using Phone”. Trained on a dataset of 2,442 annotated images, lightweight YOLOv8 variants (YOLOv8n, YOLOv8s, and YOLOv8m) achieved robust performance, with YOLOv8m attaining an mAP@50 of 0.962 and mAP@50:95 of 0.790. A Streamlit-based web application was developed to facilitate real-time behavior monitoring, enhancing usability for educators and providing insights into teaching effectiveness. This work demonstrates the potential of AI-driven systems to transform classroom management and support data-informed pedagogical strategies.

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

Real-Time Student Behavior Detection in Classroom Using YOLOv8 Model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Quang-Thuan Nguyen,
  • Dinh-Thai Kim,
  • Xuan-Hai Le,
  • Anh-Phuong Pham

摘要

Effective classroom management relies on accurate monitoring of student behaviors, yet traditional manual observation is subjective and labor-intensive. This study leverages YOLOv8, a state-of-the-art object detection framework, to automatically detect behaviors such as “Focused”, “Raising Hand”, “Distracted”, “Sleep”, and “Using Phone”. Trained on a dataset of 2,442 annotated images, lightweight YOLOv8 variants (YOLOv8n, YOLOv8s, and YOLOv8m) achieved robust performance, with YOLOv8m attaining an mAP@50 of 0.962 and mAP@50:95 of 0.790. A Streamlit-based web application was developed to facilitate real-time behavior monitoring, enhancing usability for educators and providing insights into teaching effectiveness. This work demonstrates the potential of AI-driven systems to transform classroom management and support data-informed pedagogical strategies.