Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector

Research output: Contribution to conferencePaper

16 Citations (Scopus)
55 Downloads (Pure)

Abstract

In the security domain a key problem is identifying rare behaviours of interest. Training examples for these behaviours may or may not exist, and if they do exist there will be few examples, quite probably one. We present a novel weakly supervised algorithm that can detect behaviours that either have never before been seen or for which there are few examples. Global context is modelled, allowing the detection of abnormal behaviours that in isolation appear normal. Pragmatic aspects are considered, such that no parameter tuning is required and real time performance is achieved.
Original languageEnglish
Pages2198-2205
Number of pages8
Publication statusPublished - 2011
Event13th International Conference on Computer Vision (ICCV) - Barcelona, Spain
Duration: 6 Nov 201113 Nov 2011

Conference

Conference13th International Conference on Computer Vision (ICCV)
CountrySpain
CityBarcelona
Period6/11/1113/11/11

Fingerprint

Tuning
Detectors

Cite this

Fincham Haines, T., & Xiang, T. (2011). Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector. 2198-2205. Paper presented at 13th International Conference on Computer Vision (ICCV), Barcelona, Spain.

Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector. / Fincham Haines, Tom; Xiang, Tao.

2011. 2198-2205 Paper presented at 13th International Conference on Computer Vision (ICCV), Barcelona, Spain.

Research output: Contribution to conferencePaper

Fincham Haines, T & Xiang, T 2011, 'Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector' Paper presented at 13th International Conference on Computer Vision (ICCV), Barcelona, Spain, 6/11/11 - 13/11/11, pp. 2198-2205.
Fincham Haines T, Xiang T. Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector. 2011. Paper presented at 13th International Conference on Computer Vision (ICCV), Barcelona, Spain.
Fincham Haines, Tom ; Xiang, Tao. / Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector. Paper presented at 13th International Conference on Computer Vision (ICCV), Barcelona, Spain.8 p.
@conference{7eef50c15c274dbca710bd248578015d,
title = "Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector",
abstract = "In the security domain a key problem is identifying rare behaviours of interest. Training examples for these behaviours may or may not exist, and if they do exist there will be few examples, quite probably one. We present a novel weakly supervised algorithm that can detect behaviours that either have never before been seen or for which there are few examples. Global context is modelled, allowing the detection of abnormal behaviours that in isolation appear normal. Pragmatic aspects are considered, such that no parameter tuning is required and real time performance is achieved.",
author = "{Fincham Haines}, Tom and Tao Xiang",
year = "2011",
language = "English",
pages = "2198--2205",
note = "13th International Conference on Computer Vision (ICCV) ; Conference date: 06-11-2011 Through 13-11-2011",

}

TY - CONF

T1 - Delta-dual hierarchical Dirichlet processes: A pragmatic abnormal behaviour detector

AU - Fincham Haines, Tom

AU - Xiang, Tao

PY - 2011

Y1 - 2011

N2 - In the security domain a key problem is identifying rare behaviours of interest. Training examples for these behaviours may or may not exist, and if they do exist there will be few examples, quite probably one. We present a novel weakly supervised algorithm that can detect behaviours that either have never before been seen or for which there are few examples. Global context is modelled, allowing the detection of abnormal behaviours that in isolation appear normal. Pragmatic aspects are considered, such that no parameter tuning is required and real time performance is achieved.

AB - In the security domain a key problem is identifying rare behaviours of interest. Training examples for these behaviours may or may not exist, and if they do exist there will be few examples, quite probably one. We present a novel weakly supervised algorithm that can detect behaviours that either have never before been seen or for which there are few examples. Global context is modelled, allowing the detection of abnormal behaviours that in isolation appear normal. Pragmatic aspects are considered, such that no parameter tuning is required and real time performance is achieved.

M3 - Paper

SP - 2198

EP - 2205

ER -