Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance

Video surveillance is widely used in different areas like roads, malls, education, industries, retail, parks, bus stands, and restaurants, each presenting distinct anomaly patterns that demand specialized detection strategies. Adapting anomaly detection models to new camera viewpoints or environment...

Full description

Saved in:
Bibliographic Details
Main Authors: Deepak Kumar Singh, Dibakar Raj Pant, Ganesh Gautam, Bhanu Shrestha
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/12/6687
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849432167351844864
author Deepak Kumar Singh
Dibakar Raj Pant
Ganesh Gautam
Bhanu Shrestha
author_facet Deepak Kumar Singh
Dibakar Raj Pant
Ganesh Gautam
Bhanu Shrestha
author_sort Deepak Kumar Singh
collection DOAJ
description Video surveillance is widely used in different areas like roads, malls, education, industries, retail, parks, bus stands, and restaurants, each presenting distinct anomaly patterns that demand specialized detection strategies. Adapting anomaly detection models to new camera viewpoints or environmental variations within the same scenario remains a significant challenge. Extending these models to entirely different surveillance environments or scenarios often requires extensive retraining, which can be both resource-intensive and time-consuming. To overcome these limitations, model frameworks, i.e., the video anomaly detector model, have been proposed, leveraging the meta-learning framework for faster adaptation using swin transformer for feature extraction to new concepts. In response, the dataset named MSAD (multi-scenario anomaly detection) having 14 different scenarios from multiple camera views, is the high resolution anomaly detection dataset that includes diverse motion patterns and challenging variations such as varying lighting and weather conditions, offering a robust foundation for training advanced anomaly detection models. Experiments validate the effectiveness of the proposed framework, which integrates model-agnostic meta-learning (MAML) with a ten-shot, one-query adaptation strategy. Leveraging the swin transformer as a spatial feature extractor, the model captures rich hierarchical representations from surveillance videos. This combination enables rapid generalization to novel viewpoints within the same scenario and maintains competitive performance when deployed in entirely new environments. These results highlight the strength of MAML in few-shot learning settings and demonstrate its potential for scalable anomaly detection across diverse surveillance scenarios.
format Article
id doaj-art-4660dcae6ad74f53bf55605452d4a1fc
institution Kabale University
issn 2076-3417
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-4660dcae6ad74f53bf55605452d4a1fc2025-08-20T03:27:26ZengMDPI AGApplied Sciences2076-34172025-06-011512668710.3390/app15126687Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video SurveillanceDeepak Kumar Singh0Dibakar Raj Pant1Ganesh Gautam2Bhanu Shrestha3Department of Electronics and Computer Engineering, Pulchowk Campus, Tribhuvan University, Kathmandu 44613, NepalDepartment of Electronics and Computer Engineering, Pulchowk Campus, Tribhuvan University, Kathmandu 44613, NepalDepartment of Electronics and Computer Engineering, Pulchowk Campus, Tribhuvan University, Kathmandu 44613, NepalDepartment of Information Convergence System, Graduate School of Smart Convergence, Kwangwoon University, Seoul 01897, Republic of KoreaVideo surveillance is widely used in different areas like roads, malls, education, industries, retail, parks, bus stands, and restaurants, each presenting distinct anomaly patterns that demand specialized detection strategies. Adapting anomaly detection models to new camera viewpoints or environmental variations within the same scenario remains a significant challenge. Extending these models to entirely different surveillance environments or scenarios often requires extensive retraining, which can be both resource-intensive and time-consuming. To overcome these limitations, model frameworks, i.e., the video anomaly detector model, have been proposed, leveraging the meta-learning framework for faster adaptation using swin transformer for feature extraction to new concepts. In response, the dataset named MSAD (multi-scenario anomaly detection) having 14 different scenarios from multiple camera views, is the high resolution anomaly detection dataset that includes diverse motion patterns and challenging variations such as varying lighting and weather conditions, offering a robust foundation for training advanced anomaly detection models. Experiments validate the effectiveness of the proposed framework, which integrates model-agnostic meta-learning (MAML) with a ten-shot, one-query adaptation strategy. Leveraging the swin transformer as a spatial feature extractor, the model captures rich hierarchical representations from surveillance videos. This combination enables rapid generalization to novel viewpoints within the same scenario and maintains competitive performance when deployed in entirely new environments. These results highlight the strength of MAML in few-shot learning settings and demonstrate its potential for scalable anomaly detection across diverse surveillance scenarios.https://www.mdpi.com/2076-3417/15/12/6687meta-learningMSAD datasetswin transformervideo anomaly detector
spellingShingle Deepak Kumar Singh
Dibakar Raj Pant
Ganesh Gautam
Bhanu Shrestha
Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
Applied Sciences
meta-learning
MSAD dataset
swin transformer
video anomaly detector
title Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
title_full Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
title_fullStr Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
title_full_unstemmed Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
title_short Meta-Learning Approach for Adaptive Anomaly Detection from Multi-Scenario Video Surveillance
title_sort meta learning approach for adaptive anomaly detection from multi scenario video surveillance
topic meta-learning
MSAD dataset
swin transformer
video anomaly detector
url https://www.mdpi.com/2076-3417/15/12/6687
work_keys_str_mv AT deepakkumarsingh metalearningapproachforadaptiveanomalydetectionfrommultiscenariovideosurveillance
AT dibakarrajpant metalearningapproachforadaptiveanomalydetectionfrommultiscenariovideosurveillance
AT ganeshgautam metalearningapproachforadaptiveanomalydetectionfrommultiscenariovideosurveillance
AT bhanushrestha metalearningapproachforadaptiveanomalydetectionfrommultiscenariovideosurveillance