Lighter student engagement recognition in a classroom environment using skeletal keypoints
Published in IAENG International Journal of Computer Science, 52(6), 1997-2014, June 2025, 2025
A lightweight approach to recognizing student engagement in classroom environments using skeletal keypoints, published in the IAENG International Journal of Computer Science. The proposed method combines YOLOv8m person detection with MediaPipe skeletal keypoint estimation, outperforming the baseline (YOLOv4 + OpenPose) on accuracy and speed. Code and companion data are in the skeletal-keypoint-engagement repository.
The full paper is available for download here.
Recommended citation: Tarigan, G. A., Elwirehardja, G. N., Nugroho, K. S., & Pardamean, B. (2025). "Lighter student engagement recognition in a classroom environment using skeletal keypoints." IAENG International Journal of Computer Science, 52(6), 1997-2014.
Download Paper