Variational Maximum A Posteriori model similarity and dissimilarity matching

John Chiverton; Majid Mirmehdi; Xianghua Xie

doi:10.1109/ICPR.2008.4761600

Variational Maximum A Posteriori model similarity and dissimilarity matching

John Chiverton, Majid Mirmehdi, Xianghua Xie

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

169 Downloads (Pure)

Abstract

A new variational Maximum A Posteriori (MAP) contextual modeling approach is presented that minimizes the product of two ratios: (a) the ratio of the model distribution to the distribution of currently estimated foreground pixels; (b) the ratio of the background distribution to the model distribution for all estimated background pixels. This approach provides robust discrimination to identify the division between foreground and background pixels, which is useful for applications such as object tracking.

Original language	English
Title of host publication	19th International conference on pattern recognition, 2008. ICPR 2008
Place of Publication	Piscataway
Publisher	IEEE/ IAPR
Pages	1-4
ISBN (Electronic)	9781424421756
ISBN (Print)	9781424421749
DOIs	https://doi.org/10.1109/ICPR.2008.4761600
Publication status	Published - 1 Dec 2008
Event	19th International conference on pattern recognition - Florida, United States Duration: 8 Dec 2008 → 11 Dec 2008

Conference

Conference	19th International conference on pattern recognition
Abbreviated title	ICPR 2008
Country/Territory	United States
City	Florida
Period	8/12/08 → 11/12/08

Access to Document

10.1109/ICPR.2008.4761600

ChivertonEtAlICPR2008Accepted author manuscript (Post-print), 1.18 MB

Cite this

@inproceedings{20dbf27cc4cf4f9c9c76b30c4fb06bf3,

title = "Variational Maximum A Posteriori model similarity and dissimilarity matching",

abstract = "A new variational Maximum A Posteriori (MAP) contextual modeling approach is presented that minimizes the product of two ratios: (a) the ratio of the model distribution to the distribution of currently estimated foreground pixels; (b) the ratio of the background distribution to the model distribution for all estimated background pixels. This approach provides robust discrimination to identify the division between foreground and background pixels, which is useful for applications such as object tracking.",

author = "John Chiverton and Majid Mirmehdi and Xianghua Xie",

year = "2008",

month = dec,

day = "1",

doi = "10.1109/ICPR.2008.4761600",

language = "English",

isbn = "9781424421749 ",

pages = "1--4",

booktitle = "19th International conference on pattern recognition, 2008. ICPR 2008",

publisher = "IEEE/ IAPR",

note = "19th International conference on pattern recognition, ICPR 2008 ; Conference date: 08-12-2008 Through 11-12-2008",

}

TY - GEN

T1 - Variational Maximum A Posteriori model similarity and dissimilarity matching

AU - Chiverton, John

AU - Mirmehdi, Majid

AU - Xie, Xianghua

PY - 2008/12/1

Y1 - 2008/12/1

N2 - A new variational Maximum A Posteriori (MAP) contextual modeling approach is presented that minimizes the product of two ratios: (a) the ratio of the model distribution to the distribution of currently estimated foreground pixels; (b) the ratio of the background distribution to the model distribution for all estimated background pixels. This approach provides robust discrimination to identify the division between foreground and background pixels, which is useful for applications such as object tracking.

AB - A new variational Maximum A Posteriori (MAP) contextual modeling approach is presented that minimizes the product of two ratios: (a) the ratio of the model distribution to the distribution of currently estimated foreground pixels; (b) the ratio of the background distribution to the model distribution for all estimated background pixels. This approach provides robust discrimination to identify the division between foreground and background pixels, which is useful for applications such as object tracking.

U2 - 10.1109/ICPR.2008.4761600

DO - 10.1109/ICPR.2008.4761600

M3 - Conference contribution

SN - 9781424421749

SP - 1

EP - 4

BT - 19th International conference on pattern recognition, 2008. ICPR 2008

PB - IEEE/ IAPR

CY - Piscataway

T2 - 19th International conference on pattern recognition

Y2 - 8 December 2008 through 11 December 2008

ER -

Variational Maximum A Posteriori model similarity and dissimilarity matching

Abstract

Conference

Access to Document

Fingerprint

Cite this