A temporal–spatial deep learning framework leveraging dynamic 3D attention maps for violence detection
Neural Computing and Applications · 2025-11 · 37 (32) · 26689-26709
DOI: 10.1007/s00521-025-11641-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neural Computing and Applications, online erschienen am 16.09.2025. Verfasst von Elizabeth B. Varghese und 3 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 70 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qatar National Research Fund, Hamad bin Khalifa University.
Abstract (Originalsprache)
In intelligent systems for real-time security and safety monitoring, the proliferation of surveillance cameras has fueled a growing interest in using deep learning-based artificial intelligence (AI) models for violence detection. Most current approaches consider violence detection as a video classification task, overlooking the fact that violent activities occur within relatively small spatiotemporal regions. Moreover, these activities depend on relationships among multiple such regions, making a single region analysis inadequate, especially for larger-scale violence. This paper proposes a novel temporal–spatial attention framework inspired by human visual perception, which dynamically focuses on multiple informative regions across space and time. By learning where, when, and for how long to attend within a video, using dynamic three-dimensional attention prediction networks, the model captures complex patterns of violent behavior more effectively. Experiments on four public benchmark datasets and a real-world dataset created for this study demonstrate that the proposed approach outperforms existing methods in accuracy and interpretability.
Förderung
- Qatar National Research Fund
- Hamad bin Khalifa University
Zitieren
APA
Varghese, E. B., Elzein, A., Yang, Y., & Qaraqe, M. (2025). A temporal–spatial deep learning framework leveraging dynamic 3D attention maps for violence detection. Neural Computing and Applications, 37(32), 26689-26709. https://doi.org/10.1007/s00521-025-11641-4
Vancouver
Varghese EB, Elzein A, Yang Y, Qaraqe M. A temporal–spatial deep learning framework leveraging dynamic 3D attention maps for violence detection. Neural Computing and Applications. 2025;37(32):26689-26709. doi:10.1007/s00521-025-11641-4
BibTeX
@article{Varghese2025,
author = {Varghese, Elizabeth B. and Elzein, Almiqdad and Yang, Yin and Qaraqe, Marwa},
title = {A temporal–spatial deep learning framework leveraging dynamic 3D attention maps for violence detection},
journal = {Neural Computing and Applications},
year = {2025},
volume = {37},
number = {32},
pages = {26689--26709},
doi = {10.1007/s00521-025-11641-4}
}
EUVIMED