Running summarizing tools on a new article

Check the paper below, then start its synthesis.


Intelligent Classroom Student Behavior Recognition Based on Enhanced YOLOv8

Xinhui Wang (Qingdao City University), Gao Y (Qingdao City University), Yunfei Zhang (Qingdao City University), Lei Du (Qingdao City University)

DOI: 10.1145/3797552.3797731 (Computer Science)
License: CC BY

Abstract: Accurate recognition of student classroom behaviors in intelligent teaching environments is crucial for improving teaching quality and learning effectiveness. To address technical challenges in real classroom scenarios, such as large-scale target variations, severe occlusions, and dense distributions, this paper proposes an enhanced YOLOv8-based student behavior recognition framework. First, by integrating the multi-scale feature extraction capability of Res2Net with the efficient detection structure of YOLOv8, we design a Multi-Scale Feature Fusion Block (MSF-Block) to enhance multi-scale feature learning capabilities. Second, we integrate an Efficient Channel-Spatial Attention mechanism (ECSA) into the backbone network to improve feature representation capabilities and introduce a Cross-Scale Adaptive Fusion module (CSAF) into the neck network to enhance contextual information capture. Experimental results on the SCB dataset demonstrate that compared to the original YOLOv8 model, our proposed method achieves a 4.3% improvement in mean Average Precision ([email protected]), reaching 76.4%, while maintaining good real-time performance.

Published on 21 Nov. 2025