Background subtraction with Dirichlet processes

Research output: Contribution to conferencePaper

28 Citations (Scopus)
66 Downloads (Pure)

Abstract

Background subtraction is an important first step for video analysis, where it is used to discover the objects of interest for further processing. Such an algorithm often consists of a background model and a regularisation scheme. The background model determines a per-pixel measure of if a pixel belongs to the background or the foreground, whilst the regularisation brings in information from adjacent pixels. A new method is presented that uses a Dirichlet process Gaussian mixture model to estimate a per-pixel background distribution, which is followed by probabilistic regularisation. Key advantages include inferring the per-pixel mode count, such that it accurately models dynamic backgrounds, and that it updates its model continuously in a principled way.
Original languageEnglish
Pages99-113
Number of pages15
Publication statusPublished - 2012
Event12th European Conference on Computer Vision,2012 - Florence, Italy
Duration: 7 Oct 201213 Oct 2012

Conference

Conference12th European Conference on Computer Vision,2012
CountryItaly
CityFlorence
Period7/10/1213/10/12

Fingerprint

Pixels
Dynamic models
Processing

Cite this

Fincham Haines, T., & Xiang, T. (2012). Background subtraction with Dirichlet processes. 99-113. Paper presented at 12th European Conference on Computer Vision,2012, Florence, Italy.

Background subtraction with Dirichlet processes. / Fincham Haines, Tom; Xiang, Tao.

2012. 99-113 Paper presented at 12th European Conference on Computer Vision,2012, Florence, Italy.

Research output: Contribution to conferencePaper

Fincham Haines, T & Xiang, T 2012, 'Background subtraction with Dirichlet processes' Paper presented at 12th European Conference on Computer Vision,2012, Florence, Italy, 7/10/12 - 13/10/12, pp. 99-113.
Fincham Haines T, Xiang T. Background subtraction with Dirichlet processes. 2012. Paper presented at 12th European Conference on Computer Vision,2012, Florence, Italy.
Fincham Haines, Tom ; Xiang, Tao. / Background subtraction with Dirichlet processes. Paper presented at 12th European Conference on Computer Vision,2012, Florence, Italy.15 p.
@conference{0a8a0060b6284981a1581b450a82957c,
title = "Background subtraction with Dirichlet processes",
abstract = "Background subtraction is an important first step for video analysis, where it is used to discover the objects of interest for further processing. Such an algorithm often consists of a background model and a regularisation scheme. The background model determines a per-pixel measure of if a pixel belongs to the background or the foreground, whilst the regularisation brings in information from adjacent pixels. A new method is presented that uses a Dirichlet process Gaussian mixture model to estimate a per-pixel background distribution, which is followed by probabilistic regularisation. Key advantages include inferring the per-pixel mode count, such that it accurately models dynamic backgrounds, and that it updates its model continuously in a principled way.",
author = "{Fincham Haines}, Tom and Tao Xiang",
year = "2012",
language = "English",
pages = "99--113",
note = "12th European Conference on Computer Vision,2012 ; Conference date: 07-10-2012 Through 13-10-2012",

}

TY - CONF

T1 - Background subtraction with Dirichlet processes

AU - Fincham Haines, Tom

AU - Xiang, Tao

PY - 2012

Y1 - 2012

N2 - Background subtraction is an important first step for video analysis, where it is used to discover the objects of interest for further processing. Such an algorithm often consists of a background model and a regularisation scheme. The background model determines a per-pixel measure of if a pixel belongs to the background or the foreground, whilst the regularisation brings in information from adjacent pixels. A new method is presented that uses a Dirichlet process Gaussian mixture model to estimate a per-pixel background distribution, which is followed by probabilistic regularisation. Key advantages include inferring the per-pixel mode count, such that it accurately models dynamic backgrounds, and that it updates its model continuously in a principled way.

AB - Background subtraction is an important first step for video analysis, where it is used to discover the objects of interest for further processing. Such an algorithm often consists of a background model and a regularisation scheme. The background model determines a per-pixel measure of if a pixel belongs to the background or the foreground, whilst the regularisation brings in information from adjacent pixels. A new method is presented that uses a Dirichlet process Gaussian mixture model to estimate a per-pixel background distribution, which is followed by probabilistic regularisation. Key advantages include inferring the per-pixel mode count, such that it accurately models dynamic backgrounds, and that it updates its model continuously in a principled way.

M3 - Paper

SP - 99

EP - 113

ER -