Enhancing Video Anomaly Detection Using a Transformer Spatiotemporal Attention Unsupervised Framework for Large Datasets
Algorithms · 01.07.2024 · 17 (7) · 286
DOI: 10.3390/a17070286 ↗Kurz gefasst
Fachartikel in der Zeitschrift Algorithms, online erschienen am 01.07.2024. Verfasst von Mohamed H. Habeb und 2 weiteren Autor:innen. Die Arbeit wurde bisher 31-mal zitiert (Crossref) und nennt 95 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Information Technology Industry Development Agency (ITIDA)–Information Technology Academia Collaboration (ITAC).
Abstract (Originalsprache)
This work introduces an unsupervised framework for video anomaly detection, leveraging a hybrid deep learning model that combines a vision transformer (ViT) with a convolutional spatiotemporal relationship (STR) attention block. The proposed model addresses the challenges of anomaly detection in video surveillance by capturing both local and global relationships within video frames, a task that traditional convolutional neural networks (CNNs) often struggle with due to their localized field of view. We have utilized a pre-trained ViT as an encoder for feature extraction, which is then processed by the STR attention block to enhance the detection of spatiotemporal relationships among objects in videos. The novelty of this work is utilizing the ViT with the STR attention to detect video anomalies effectively in large and heterogeneous datasets, an important thing given the diverse environments and scenarios encountered in real-world surveillance. The framework was evaluated on three benchmark datasets, i.e., the UCSD-Ped2, CHUCK Avenue, and ShanghaiTech. This demonstrates the model’s superior performance in detecting anomalies compared to state-of-the-art methods, showcasing its potential to significantly enhance automated video surveillance systems by achieving area under the receiver operating characteristic curve (AUC ROC) values of 95.6, 86.8, and 82.1. To show the effectiveness of the proposed framework in detecting anomalies in extra-large datasets, we trained the model on a subset of the huge contemporary CHAD dataset that contains over 1 million frames, achieving AUC ROC values of 71.8 and 64.2 for CHAD-Cam 1 and CHAD-Cam 2, respectively, which outperforms the state-of-the-art techniques.
Förderung
- Information Technology Industry Development Agency (ITIDA)–Information Technology Academia Collaboration (ITAC)
Zitieren
APA
Habeb, M. H., Salama, M., & Elrefaei, L. A. (2024). Enhancing Video Anomaly Detection Using a Transformer Spatiotemporal Attention Unsupervised Framework for Large Datasets. Algorithms, 17(7), 286. https://doi.org/10.3390/a17070286
Vancouver
Habeb MH, Salama M, Elrefaei LA. Enhancing Video Anomaly Detection Using a Transformer Spatiotemporal Attention Unsupervised Framework for Large Datasets. Algorithms. 2024;17(7):286. doi:10.3390/a17070286
BibTeX
@article{Habeb2024,
author = {Habeb, Mohamed H. and Salama, May and Elrefaei, Lamiaa A.},
title = {Enhancing Video Anomaly Detection Using a Transformer Spatiotemporal Attention Unsupervised Framework for Large Datasets},
journal = {Algorithms},
year = {2024},
volume = {17},
number = {7},
pages = {286},
doi = {10.3390/a17070286}
}
EUVIMED