EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CSSA-YOLO: Cross-Scale Spatiotemporal Attention Network for Fine-Grained Behavior Recognition in Classroom Environments

Liuchen Zhou iD, Xiangpeng Liu iD, Xiqiang Guan iD, Yuhua Cheng iD

Sensors · 15.05.2025 · 25 (10) · 3132

DOI: 10.3390/s25103132 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 15.05.2025. Verfasst von Liuchen Zhou und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanghai Normal University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Under a student-centered educational paradigm, project-based learning (PBL) assessment requires accurate identification of classroom behaviors to facilitate effective teaching evaluations and the implementation of personalized learning strategies. The increasing use of visual and multi-modal sensors in smart classrooms has made it possible to continuously capture rich behavioral data. However, challenges such as lighting variations, occlusions, and diverse behaviors complicate sensor-based behavior analysis. To address these issues, we introduce CSSA-YOLO, a novel detection network that incorporates cross-scale feature optimization. First, we establish a C2fs module that captures spatiotemporal dependencies in small-scale actions such as hand-raising through hierarchical window attention. Second, a Shuffle Attention mechanism is then integrated into the neck to suppress interference from complex backgrounds, thereby enhancing the model’s ability to focus on relevant features. Finally, to further enhance the network’s ability to detect small targets and complex boundary behaviors, we utilize the WIoU loss function, which dynamically weights gradients to optimize the localization accuracy of occluded targets. Experiments involving the SCB03-S dataset showed that CSSA-YOLO outperforms traditional methods, achieving an mAP50 of 76.0%, surpassing YOLOv8m by 1.2%, particularly in complex background and occlusion scenarios. Furthermore, it reaches 78.31 FPS, meeting the requirements for real-time application. This study offers a reliable solution for precise behavior recognition in classroom settings, supporting the development of intelligent education systems.

Schlagwörter

Stichwörter der Autor:innen

Shuffle AttentionSwin TransformerWIoU losscross-scale feature refinementstudent behavior recognition

Autor:innen und Institutionen

  • College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China.
  • Shanghai Research Institute of Microelectronics, Peking University, Shanghai 201203, China.

ORCID-Profile: Liuchen Zhou, Xiangpeng Liu, Xiqiang Guan, Yuhua Cheng

Förderung

  • Shanghai Normal University

Zitieren

APA

Zhou, L., Liu, X., Guan, X., & Cheng, Y. (2025). CSSA-YOLO: Cross-Scale Spatiotemporal Attention Network for Fine-Grained Behavior Recognition in Classroom Environments. Sensors, 25(10), 3132. https://doi.org/10.3390/s25103132

Vancouver

Zhou L, Liu X, Guan X, Cheng Y. CSSA-YOLO: Cross-Scale Spatiotemporal Attention Network for Fine-Grained Behavior Recognition in Classroom Environments. Sensors. 2025;25(10):3132. doi:10.3390/s25103132

BibTeX
@article{Zhou2025,
  author = {Zhou, Liuchen and Liu, Xiangpeng and Guan, Xiqiang and Cheng, Yuhua},
  title = {CSSA-YOLO: Cross-Scale Spatiotemporal Attention Network for Fine-Grained Behavior Recognition in Classroom Environments},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {10},
  pages = {3132},
  doi = {10.3390/s25103132}
}
Kostenlos starten →