Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation

Detecting anomaly patterns in videos is a challenging task due to complex scenes, huge diversity of anomalies, and fuzzy nature of the task. With advent of technology, tremendous size of visual data is being generated by video surveillance systems, which makes harder to search, analyze, and detect a...

Full description

Saved in:
Bibliographic Details
Main Authors: Burcak Asal, Ahmet Burak Can
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11071705/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849427436664520704
author Burcak Asal
Ahmet Burak Can
author_facet Burcak Asal
Ahmet Burak Can
author_sort Burcak Asal
collection DOAJ
description Detecting anomaly patterns in videos is a challenging task due to complex scenes, huge diversity of anomalies, and fuzzy nature of the task. With advent of technology, tremendous size of visual data is being generated by video surveillance systems, which makes harder to search, analyze, and detect anomalies on video data by human operators. In this paper, we introduce three relational distillation approaches to handle both robust detection of anomalous events and gradual adaptation to different anomaly patterns in new videos while not forgetting anomaly patterns learned from the previous video data. In order to realize these concepts, we propose a unique attention mechanism with feature and relation based knowledge distillation methods. We adapted our knowledge distillation methods to two state-of-the-art models designed for anomaly detection task. Our extensive experiments on two public datasets show that not only our best version model achieves robust performance with a frame-level AUC of 80.22 on UCF-Crime and video-level AUC of 78.20 on RWF-2000 datasets but also the proposed distillation methods improve the performance while reducing catastrophic forgetting problem.
format Article
id doaj-art-a3766471a76943b6b2dfc3d7d284b7b5
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-a3766471a76943b6b2dfc3d7d284b7b52025-08-20T03:29:02ZengIEEEIEEE Access2169-35362025-01-011311717011718510.1109/ACCESS.2025.358598411071705Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge DistillationBurcak Asal0https://orcid.org/0009-0003-3729-8170Ahmet Burak Can1https://orcid.org/0000-0002-0101-6878Department of Computer Engineering, Adana Alparslan Türkeş Science and Technology University, Adana, TürkiyeDepartment of Computer Engineering, Hacettepe University, Ankara, TürkiyeDetecting anomaly patterns in videos is a challenging task due to complex scenes, huge diversity of anomalies, and fuzzy nature of the task. With advent of technology, tremendous size of visual data is being generated by video surveillance systems, which makes harder to search, analyze, and detect anomalies on video data by human operators. In this paper, we introduce three relational distillation approaches to handle both robust detection of anomalous events and gradual adaptation to different anomaly patterns in new videos while not forgetting anomaly patterns learned from the previous video data. In order to realize these concepts, we propose a unique attention mechanism with feature and relation based knowledge distillation methods. We adapted our knowledge distillation methods to two state-of-the-art models designed for anomaly detection task. Our extensive experiments on two public datasets show that not only our best version model achieves robust performance with a frame-level AUC of 80.22 on UCF-Crime and video-level AUC of 78.20 on RWF-2000 datasets but also the proposed distillation methods improve the performance while reducing catastrophic forgetting problem.https://ieeexplore.ieee.org/document/11071705/AR-Netcomputer visionGCNknowledge distillationrelational approachesvideo anomaly detection
spellingShingle Burcak Asal
Ahmet Burak Can
Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
IEEE Access
AR-Net
computer vision
GCN
knowledge distillation
relational approaches
video anomaly detection
title Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
title_full Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
title_fullStr Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
title_full_unstemmed Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
title_short Adaptive Video Anomaly Detection by Attention-Based Relational Knowledge Distillation
title_sort adaptive video anomaly detection by attention based relational knowledge distillation
topic AR-Net
computer vision
GCN
knowledge distillation
relational approaches
video anomaly detection
url https://ieeexplore.ieee.org/document/11071705/
work_keys_str_mv AT burcakasal adaptivevideoanomalydetectionbyattentionbasedrelationalknowledgedistillation
AT ahmetburakcan adaptivevideoanomalydetectionbyattentionbasedrelationalknowledgedistillation