Deakin University
Browse

File(s) under permanent embargo

Learning regularity in skeleton trajectories for anomaly detection in videos

Version 2 2024-06-03, 06:49
Version 1 2020-02-20, 12:33
conference contribution
posted on 2024-06-03, 06:49 authored by R Morais, V Le, Truyen TranTruyen Tran, B Saha, M Mansour, Svetha VenkateshSvetha Venkatesh
© 2019 IEEE. Appearance features have been widely used in video anomaly detection even though they contain complex entangled factors. We propose a new method to model the normal patterns of human movements in surveillance video for anomaly detection using dynamic skeleton features. We decompose the skeletal movements into two sub-components: global body movement and local body posture. We model the dynamics and interaction of the coupled features in our novel Message-Passing Encoder-Decoder Recurrent Network. We observed that the decoupled features collaboratively interact in our spatio-temporal model to accurately identify human-related irregular events from surveillance video sequences. Compared to traditional appearance-based models, our method achieves superior outlier detection performance. Our model also offers 'open-box' examination and decision explanation made possible by the semantically understandable features and a network architecture supporting interpretability.

History

Pagination

11988-11996

Location

Long Beach, California

Start date

2019-06-15

End date

2019-06-20

ISSN

1063-6919

ISBN-13

9781728132938

Language

eng

Publication classification

E1 Full written paper - refereed

Title of proceedings

CVPR 2019 : Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

Event

Computer Vision and Pattern Recognition. Conference (2019 : Long Beach, California)

Publisher

IEEE

Place of publication

Piscataway, N.J.

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC