Temporal Improvement of Video Traffic Anomaly Detection: A Positioning Paper

Xinyue Zhang, Yuxuan Zhao, Ka Lok Man, Jeremy S. Smith, Young Ae Jung, Yutao Yue*

*Corresponding author for this work

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

Abstract

Internet of Things (IoT) technology can provide real-time public facilities data to Intelligent Transportation Systems (ITS), especially the surveillance cameras' video data. With the popularity of video surveillance systems, Video Anomaly Detection (VAD) has become more important in society and traffic management. However, the traditional, highly manual-dependent VAD method and the neural network model-based VAD method (such as Recurrent Neural Networks, RNN) face a significant challenge in temporal stream processing. This paper analyses current specific temporal challenges and proposes a Transformer-based spatial-temporal VAD model to alleviate the influence of temporal limitations. With the development of Transformer models, global information consideration and long-term relationship building have become more accessible, and the processing of temporal information in video data has become more efficient.

Original languageEnglish
Title of host publication2024 International Conference on Platform Technology and Service, PlatCon 2024 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages155-159
Number of pages5
ISBN (Electronic)9798350367874
DOIs
Publication statusPublished - 2024
Event10th International Conference on Platform Technology and Service, PlatCon 2024 - Jeju, Korea, Republic of
Duration: 26 Aug 202428 Aug 2024

Publication series

Name2024 International Conference on Platform Technology and Service, PlatCon 2024 - Proceedings

Conference

Conference10th International Conference on Platform Technology and Service, PlatCon 2024
Country/TerritoryKorea, Republic of
CityJeju
Period26/08/2428/08/24

Keywords

  • Deep Learning (DL)
  • Intelligent Transportation System (ITS)
  • Video Anomaly Detection (VAD)

Fingerprint

Dive into the research topics of 'Temporal Improvement of Video Traffic Anomaly Detection: A Positioning Paper'. Together they form a unique fingerprint.

Cite this