Improved anomaly detection in crowded scenes via cell-based analysis of foreground speed, size and texture

Reddy, Vikas, Sanderson, Conrad and Lovell, Brian C. (2011). Improved anomaly detection in crowded scenes via cell-based analysis of foreground speed, size and texture. In: 2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Colorado Springs, CO, United States, (55-61). 20-25 June 2011. doi:10.1109/CVPRW.2011.5981799


Author Reddy, Vikas
Sanderson, Conrad
Lovell, Brian C.
Title of paper Improved anomaly detection in crowded scenes via cell-based analysis of foreground speed, size and texture
Conference name 2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Conference location Colorado Springs, CO, United States
Conference dates 20-25 June 2011
Proceedings title 2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)   Check publisher's open access policy
Journal name IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops   Check publisher's open access policy
Place of Publication Piscataway, NJ, United States
Publisher IEEE
Publication Year 2011
Sub-type Fully published paper
DOI 10.1109/CVPRW.2011.5981799
ISBN 9781457705298
ISSN 2160-7508
Start page 55
End page 61
Total pages 7
Collection year 2012
Language eng
Abstract/Summary A robust and efficient anomaly detection technique is proposed, capable of dealing with crowded scenes where traditional tracking based approaches tend to fail. Initial foreground segmentation of the input frames confines the analysis to foreground objects and effectively ignores irrelevant background dynamics. Input frames are split into nonoverlapping cells, followed by extracting features based on motion, size and texture from each cell. Each feature type is independently analysed for the presence of an anomaly. Unlike most methods, a refined estimate of object motion is achieved by computing the optical flow of only the foreground pixels. The motion and size features are modelled by an approximated version of kernel density estimation, which is computationally efficient even for large training datasets. Texture features are modelled by an adaptively grown codebook, with the number of entries in the codebook selected in an online fashion. Experiments on the recently published UCSD Anomaly Detection dataset show that the proposed method obtains considerably better results than three recent approaches: MPPCA, social force, and mixture of dynamic textures (MDT). The proposed method is also several orders of magnitude faster than MDT, the next best performing method.
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 3 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Wed, 21 Dec 2011, 15:19:22 EST by Ms Deborah Brian on behalf of School of Information Technol and Elec Engineering