Enhancing classroom engagement using real-time student attention monitoring with YOLOv9
This research investigates the utilization of YOLOv9 in classrooms to monitor student attention in real time, thereby addressing the challenge of sustaining student engagement. The inability of conventional teaching methods to maintain students' attention frequently necessitates the implementat...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Proceeding Paper |
| Language: | English English |
| Published: |
IEEE
2024
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/115857/7/115857_%20Enhancing%20classroom%20engagement.pdf http://irep.iium.edu.my/115857/8/115857_%20Enhancing%20classroom%20engagement_Scopus.pdf http://irep.iium.edu.my/115857/ https://ieeexplore.ieee.org/document/10675573 https://doi.org/10.1109/ICSIMA62563.2024.10675573 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research investigates the utilization of YOLOv9 in classrooms to monitor student attention in real time, thereby addressing the challenge of sustaining student engagement. The inability of conventional teaching methods to maintain students' attention frequently necessitates the implementation of dynamic modifications to instructional strategies. The goal is to create a system that accurately identifies students' attention levels. The system was trained and tested on annotated datasets using the YOLOv9c and YOLOv9e models, resulting in precision, recall, and mAP scores that demonstrated high effectiveness. A dashboard from Streamlit offered real-time and cumulative insights, which allowed for immediate pedagogical adjustments. The results showed that YOLOv9c had a higher precision (0.876) and YOLOv9e had a superior recall, underscoring their respective strengths. However, despite a few misclassifications, the system's robust performance suggests its potential as a transformative educational tool. Future enhancements will focus on integrating additional data sources, reducing misclassifications, and improving algorithm scalability and accuracy to foster a more adaptive learning environment and showcase the system's potential for diverse educational applications. |
|---|