A MAP approach to landmarking

G.M. Beumer

A MAP approach to landmarking

G.M. Beumer

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

22 Downloads (Pure)

Abstract

Landmarking can be seen as calculating the maximum a posteriori probability of a certain set of landmarks given a certain image (texture) which contains a face. The MLLL plus BILBO algorithm by Beumer et al [1] is one method. An improvement to this existing landmarking algorithm, MLLL in combination with BILBO, is presented. MLLL uses a likelihood ratio based similarity score to mark candidate landmark locations. BILBO uses a statistical method on the shape to correct outliers. A theoretical analysis of this intuitive approach shows new insight. In order to verify this theory we performed a simple experiment. The results show that using the new method performs significantly better than when only using the similarity score method (MLLL) in combination with outlier removal (BILBO).

Original language	Undefined
Title of host publication	Proceedings of the 28th Symposium on Information Theory in the Benelux
Editors	Raymond N.J. Veldhuis, R.N.J. Veldhuis, H.S. Cronie
Place of Publication	Eindhoven
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)
Pages	183-188
Number of pages	6
ISBN (Print)	978-90-365-2509-1
Publication status	Published - 24 May 2007
Event	28th Symposium on Information Theory in the Benelux 2007 - Best Western Dish Hotel, Enschede, Netherlands Duration: 24 May 2007 → 25 May 2007 Conference number: 28

Publication series

Name
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie
Number	LNCS4549

Conference

Conference	28th Symposium on Information Theory in the Benelux 2007
Country	Netherlands
City	Enschede
Period	24/05/07 → 25/05/07

Keywords

EWI-10854
IR-64278
METIS-241831
SCS-Safety

Access to Document

23_05_beumer
open
23_05_beumer
open

Cite this

@inproceedings{fe6e0522839740158da64a8dd470a642,

title = "A MAP approach to landmarking",

abstract = "Landmarking can be seen as calculating the maximum a posteriori probability of a certain set of landmarks given a certain image (texture) which contains a face. The MLLL plus BILBO algorithm by Beumer et al [1] is one method. An improvement to this existing landmarking algorithm, MLLL in combination with BILBO, is presented. MLLL uses a likelihood ratio based similarity score to mark candidate landmark locations. BILBO uses a statistical method on the shape to correct outliers. A theoretical analysis of this intuitive approach shows new insight. In order to verify this theory we performed a simple experiment. The results show that using the new method performs significantly better than when only using the similarity score method (MLLL) in combination with outlier removal (BILBO).",

keywords = "EWI-10854, IR-64278, METIS-241831, SCS-Safety",

author = "G.M. Beumer",

year = "2007",

month = may,

day = "24",

language = "Undefined",

isbn = "978-90-365-2509-1",

publisher = "Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)",

number = "LNCS4549",

pages = "183--188",

editor = "Veldhuis, {Raymond N.J.} and R.N.J. Veldhuis and H.S. Cronie",

booktitle = "Proceedings of the 28th Symposium on Information Theory in the Benelux",

address = "Netherlands",

note = "null ; Conference date: 24-05-2007 Through 25-05-2007",

}

A MAP approach to landmarking. / Beumer, G.M.

Proceedings of the 28th Symposium on Information Theory in the Benelux. ed. / Raymond N.J. Veldhuis; R.N.J. Veldhuis; H.S. Cronie. Eindhoven : Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2007. p. 183-188.

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

TY - GEN

T1 - A MAP approach to landmarking

AU - Beumer, G.M.

N1 - Conference code: 28

PY - 2007/5/24

Y1 - 2007/5/24

N2 - Landmarking can be seen as calculating the maximum a posteriori probability of a certain set of landmarks given a certain image (texture) which contains a face. The MLLL plus BILBO algorithm by Beumer et al [1] is one method. An improvement to this existing landmarking algorithm, MLLL in combination with BILBO, is presented. MLLL uses a likelihood ratio based similarity score to mark candidate landmark locations. BILBO uses a statistical method on the shape to correct outliers. A theoretical analysis of this intuitive approach shows new insight. In order to verify this theory we performed a simple experiment. The results show that using the new method performs significantly better than when only using the similarity score method (MLLL) in combination with outlier removal (BILBO).

AB - Landmarking can be seen as calculating the maximum a posteriori probability of a certain set of landmarks given a certain image (texture) which contains a face. The MLLL plus BILBO algorithm by Beumer et al [1] is one method. An improvement to this existing landmarking algorithm, MLLL in combination with BILBO, is presented. MLLL uses a likelihood ratio based similarity score to mark candidate landmark locations. BILBO uses a statistical method on the shape to correct outliers. A theoretical analysis of this intuitive approach shows new insight. In order to verify this theory we performed a simple experiment. The results show that using the new method performs significantly better than when only using the similarity score method (MLLL) in combination with outlier removal (BILBO).

KW - EWI-10854

KW - IR-64278

KW - METIS-241831

KW - SCS-Safety

M3 - Conference contribution

SN - 978-90-365-2509-1

SP - 183

EP - 188

BT - Proceedings of the 28th Symposium on Information Theory in the Benelux

A2 - Veldhuis, Raymond N.J.

A2 - Veldhuis, R.N.J.

A2 - Cronie, H.S.

PB - Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)

CY - Eindhoven

Y2 - 24 May 2007 through 25 May 2007

ER -