Video Anomaly Detection for Surveillance Based on Effective Frame Area

Yuxing Yang, Yang Xian, Zeyu Fu, Syed Mohsen Naqvi

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

10 Citations (Scopus)

Abstract

Video anomaly detection aims to recognise and analyse the video sequences to classify the normal and abnormal frames. This technology can efficiently reduce the human labour to discover the anomalies in surveillance systems and is widely applied in financial, public security and transport sectors. However, video anomaly detection performance is often degraded by the dataset quality, especially for small objects in video sequences. Besides, the computational cost of the classification model would be required as low as possible. In this paper, we proposed information fusion with a joint model which contains motion estimation, object detection and adversarial learning to detect anomalies in two video datasets: UCSD PED1 and PED2. Experimental results confirm the proposed method outperforms the state-of-the-art methods with the additional advantages in reduced computation cost.

Original languageEnglish
Title of host publicationProceedings of 2021 IEEE 24th International Conference on Information Fusion, FUSION 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781737749714
Publication statusPublished - 2021
Externally publishedYes
Event24th IEEE International Conference on Information Fusion, FUSION 2021 - Sun City, South Africa
Duration: 1 Nov 20214 Nov 2021

Publication series

NameProceedings of 2021 IEEE 24th International Conference on Information Fusion, FUSION 2021

Conference

Conference24th IEEE International Conference on Information Fusion, FUSION 2021
Country/TerritorySouth Africa
CitySun City
Period1/11/214/11/21

Keywords

  • Generative Adversarial Networks
  • Motion Estimation
  • Object Detection
  • Video Anomaly Detection

Fingerprint

Dive into the research topics of 'Video Anomaly Detection for Surveillance Based on Effective Frame Area'. Together they form a unique fingerprint.

Cite this