Active online anomaly detection using dirichlet process mixture model and Gaussian process classification

Jagannadan Varadarajan, Ramanathan Subramanian, Narendra Ahuja, Pierre Moulin, Jean Marc Odobez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a novel anomaly detection (AD) system for streaming videos. Different from prior methods that rely on unsupervised learning of clip representations, that are usually coarse in nature, and batch-mode learning, we propose the combination of two non-parametric models for our task: i) Dirichlet process mixture models (DPMM) based modeling of object motion and directions in each cell, and ii) Gaussian process based active learning paradigm involving labeling by a domain expert. Whereas conventional clip representation methods adopt quantizing only motion directions leading to a lossy, coarse representation that are inadequate, our clip representation approach results in fine grained clusters at each cell that model the scene activities (both direction and speed) more effectively. For active anomaly detection, we adapt a Gaussian Process framework to process incoming samples (video snippets) sequentially, seek labels for confusing or informative samples and and update the AD model online. Furthermore, the proposed video representation along with a novel query criterion to select informative samples for labeling that incorporates both exploration and exploitation criteria is proposed, and is found to outperform competing criteria on two challenging traffic scene datasets.

Original languageEnglish (US)
Title of host publicationProceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages615-623
Number of pages9
ISBN (Electronic)9781509048229
DOIs
StatePublished - May 11 2017
Event17th IEEE Winter Conference on Applications of Computer Vision, WACV 2017 - Santa Rosa, United States
Duration: Mar 24 2017Mar 31 2017

Publication series

NameProceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017

Other

Other17th IEEE Winter Conference on Applications of Computer Vision, WACV 2017
CountryUnited States
CitySanta Rosa
Period3/24/173/31/17

Fingerprint

Labeling
Unsupervised learning
Video streaming
Labels
Problem-Based Learning

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Varadarajan, J., Subramanian, R., Ahuja, N., Moulin, P., & Odobez, J. M. (2017). Active online anomaly detection using dirichlet process mixture model and Gaussian process classification. In Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017 (pp. 615-623). [7926657] (Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WACV.2017.74

Active online anomaly detection using dirichlet process mixture model and Gaussian process classification. / Varadarajan, Jagannadan; Subramanian, Ramanathan; Ahuja, Narendra; Moulin, Pierre; Odobez, Jean Marc.

Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 615-623 7926657 (Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Varadarajan, J, Subramanian, R, Ahuja, N, Moulin, P & Odobez, JM 2017, Active online anomaly detection using dirichlet process mixture model and Gaussian process classification. in Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017., 7926657, Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017, Institute of Electrical and Electronics Engineers Inc., pp. 615-623, 17th IEEE Winter Conference on Applications of Computer Vision, WACV 2017, Santa Rosa, United States, 3/24/17. https://doi.org/10.1109/WACV.2017.74
Varadarajan J, Subramanian R, Ahuja N, Moulin P, Odobez JM. Active online anomaly detection using dirichlet process mixture model and Gaussian process classification. In Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 615-623. 7926657. (Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017). https://doi.org/10.1109/WACV.2017.74
Varadarajan, Jagannadan ; Subramanian, Ramanathan ; Ahuja, Narendra ; Moulin, Pierre ; Odobez, Jean Marc. / Active online anomaly detection using dirichlet process mixture model and Gaussian process classification. Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 615-623 (Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017).
@inproceedings{a262bd292e5a4a719cc4a9678eb78ca9,
title = "Active online anomaly detection using dirichlet process mixture model and Gaussian process classification",
abstract = "We present a novel anomaly detection (AD) system for streaming videos. Different from prior methods that rely on unsupervised learning of clip representations, that are usually coarse in nature, and batch-mode learning, we propose the combination of two non-parametric models for our task: i) Dirichlet process mixture models (DPMM) based modeling of object motion and directions in each cell, and ii) Gaussian process based active learning paradigm involving labeling by a domain expert. Whereas conventional clip representation methods adopt quantizing only motion directions leading to a lossy, coarse representation that are inadequate, our clip representation approach results in fine grained clusters at each cell that model the scene activities (both direction and speed) more effectively. For active anomaly detection, we adapt a Gaussian Process framework to process incoming samples (video snippets) sequentially, seek labels for confusing or informative samples and and update the AD model online. Furthermore, the proposed video representation along with a novel query criterion to select informative samples for labeling that incorporates both exploration and exploitation criteria is proposed, and is found to outperform competing criteria on two challenging traffic scene datasets.",
author = "Jagannadan Varadarajan and Ramanathan Subramanian and Narendra Ahuja and Pierre Moulin and Odobez, {Jean Marc}",
year = "2017",
month = "5",
day = "11",
doi = "10.1109/WACV.2017.74",
language = "English (US)",
series = "Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "615--623",
booktitle = "Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017",
address = "United States",

}

TY - GEN

T1 - Active online anomaly detection using dirichlet process mixture model and Gaussian process classification

AU - Varadarajan, Jagannadan

AU - Subramanian, Ramanathan

AU - Ahuja, Narendra

AU - Moulin, Pierre

AU - Odobez, Jean Marc

PY - 2017/5/11

Y1 - 2017/5/11

N2 - We present a novel anomaly detection (AD) system for streaming videos. Different from prior methods that rely on unsupervised learning of clip representations, that are usually coarse in nature, and batch-mode learning, we propose the combination of two non-parametric models for our task: i) Dirichlet process mixture models (DPMM) based modeling of object motion and directions in each cell, and ii) Gaussian process based active learning paradigm involving labeling by a domain expert. Whereas conventional clip representation methods adopt quantizing only motion directions leading to a lossy, coarse representation that are inadequate, our clip representation approach results in fine grained clusters at each cell that model the scene activities (both direction and speed) more effectively. For active anomaly detection, we adapt a Gaussian Process framework to process incoming samples (video snippets) sequentially, seek labels for confusing or informative samples and and update the AD model online. Furthermore, the proposed video representation along with a novel query criterion to select informative samples for labeling that incorporates both exploration and exploitation criteria is proposed, and is found to outperform competing criteria on two challenging traffic scene datasets.

AB - We present a novel anomaly detection (AD) system for streaming videos. Different from prior methods that rely on unsupervised learning of clip representations, that are usually coarse in nature, and batch-mode learning, we propose the combination of two non-parametric models for our task: i) Dirichlet process mixture models (DPMM) based modeling of object motion and directions in each cell, and ii) Gaussian process based active learning paradigm involving labeling by a domain expert. Whereas conventional clip representation methods adopt quantizing only motion directions leading to a lossy, coarse representation that are inadequate, our clip representation approach results in fine grained clusters at each cell that model the scene activities (both direction and speed) more effectively. For active anomaly detection, we adapt a Gaussian Process framework to process incoming samples (video snippets) sequentially, seek labels for confusing or informative samples and and update the AD model online. Furthermore, the proposed video representation along with a novel query criterion to select informative samples for labeling that incorporates both exploration and exploitation criteria is proposed, and is found to outperform competing criteria on two challenging traffic scene datasets.

UR - http://www.scopus.com/inward/record.url?scp=85020186679&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020186679&partnerID=8YFLogxK

U2 - 10.1109/WACV.2017.74

DO - 10.1109/WACV.2017.74

M3 - Conference contribution

AN - SCOPUS:85020186679

T3 - Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017

SP - 615

EP - 623

BT - Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -