Real-time Anomaly Detection and Localization in Crowded Scenes
Loading...
Files
Size: 2.68 MB, File format: Adobe PDF
Date
Authors
Sabokrou, M
Fathy, M
Hoseini, M
Klette, R
Supervisor
Item type
Degree name
Journal Title
Journal ISSN
Volume Title
Publisher
IEEE
Abstract
In this paper, we propose a method for real-time anomaly
detection and localization in crowded scenes. Each video is
defined as a set of non-overlapping cubic patches, and is
described using two local and global descriptors. These
descriptors capture the video properties from different aspects.
By incorporating simple and cost-effective Gaussian
classifiers, we can distinguish normal activities and anomalies
in videos. The local and global features are based on
structure similarity between adjacent patches and the features
learned in an unsupervised way, using a sparse autoencoder.
Experimental results show that our algorithm is
comparable to a state-of-the-art procedure on UCSD ped2
and UMN benchmarks, but even more time-efficient. The
experiments confirm that our system can reliably detect and
localize anomalies as soon as they happen in a video.
Description
Keywords
Streaming media; Feature extraction; Training; Real-time systems; Gaussian distribution; Benchmark testing; Reliability
Source
2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Boston, MA, 2015, pp. 56-62.
Publisher's version
Rights statement
Copyright © 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.