Discriminative Shape Alignment

M. Loog; M. de Bruijne

doi:10.1007/978-3-642-02498-6_38

Discriminative Shape Alignment

M. Loog, M. de Bruijne

Datalogisk Institut

4 Citationer (Scopus)

Abstract

The alignment of shape data to a common mean before its subsequent processing is an ubiquitous step within the area shape analysis. Current approaches to shape analysis or, as more specifically considered in this work, shape classification perform the alignment in a fully unsupervised way, not taking into account that eventually the shapes are to be assigned to two or more different classes. This work introduces a discriminative variation to well-known Procrustes alignment and demonstrates its benefit over this classical method in shape classification tasks. The focus is on two-dimensional shapes from a two-class recognition problem.

Originalsprog	Engelsk
Titel	Information Processing in Medical Imaging
Antal sider	7
Vol/bind	5636/2009
Forlag	Springer
Publikationsdato	2009
Sider	459-466
ISBN (Trykt)	978-642-02497-9
DOI	https://doi.org/10.1007/978-3-642-02498-6_38
Status	Udgivet - 2009
Begivenhed	Information Processing in Medical Imaging 2009 (IPMI ´09) - Williamsburg, VA, USA Varighed: 5 jul. 0009 → 10 jul. 0009 Konferencens nummer: 21

Konference

Konference	Information Processing in Medical Imaging 2009 (IPMI ´09)
Nummer	21
Land/Område	USA
By	Williamsburg, VA
Periode	05/07/0009 → 10/07/0009

Navn	Lecture notes in computer science
Vol/bind	5636
ISSN	0302-9743

Adgang til dokumentet

10.1007/978-3-642-02498-6_38

Citationsformater

@inproceedings{c800cf109d3411debc73000ea68e967b,

title = "Discriminative Shape Alignment",

abstract = "The alignment of shape data to a common mean before its subsequent processing is an ubiquitous step within the area shape analysis. Current approaches to shape analysis or, as more specifically considered in this work, shape classification perform the alignment in a fully unsupervised way, not taking into account that eventually the shapes are to be assigned to two or more different classes. This work introduces a discriminative variation to well-known Procrustes alignment and demonstrates its benefit over this classical method in shape classification tasks. The focus is on two-dimensional shapes from a two-class recognition problem.",

author = "M. Loog and {de Bruijne}, M.",

year = "2009",

doi = "10.1007/978-3-642-02498-6_38",

language = "English",

isbn = "978-642-02497-9",

volume = "5636/2009",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "459--466",

booktitle = "Information Processing in Medical Imaging",

note = "Information Processing in Medical Imaging 2009 (IPMI ´09) ; Conference date: 05-07-0009 Through 10-07-0009",

}

TY - GEN

T1 - Discriminative Shape Alignment

AU - Loog, M.

AU - de Bruijne, M.

N1 - Conference code: 21

PY - 2009

Y1 - 2009

N2 - The alignment of shape data to a common mean before its subsequent processing is an ubiquitous step within the area shape analysis. Current approaches to shape analysis or, as more specifically considered in this work, shape classification perform the alignment in a fully unsupervised way, not taking into account that eventually the shapes are to be assigned to two or more different classes. This work introduces a discriminative variation to well-known Procrustes alignment and demonstrates its benefit over this classical method in shape classification tasks. The focus is on two-dimensional shapes from a two-class recognition problem.

AB - The alignment of shape data to a common mean before its subsequent processing is an ubiquitous step within the area shape analysis. Current approaches to shape analysis or, as more specifically considered in this work, shape classification perform the alignment in a fully unsupervised way, not taking into account that eventually the shapes are to be assigned to two or more different classes. This work introduces a discriminative variation to well-known Procrustes alignment and demonstrates its benefit over this classical method in shape classification tasks. The focus is on two-dimensional shapes from a two-class recognition problem.

U2 - 10.1007/978-3-642-02498-6_38

DO - 10.1007/978-3-642-02498-6_38

M3 - Article in proceedings

SN - 978-642-02497-9

VL - 5636/2009

T3 - Lecture notes in computer science

SP - 459

EP - 466

BT - Information Processing in Medical Imaging

PB - Springer

T2 - Information Processing in Medical Imaging 2009 (IPMI ´09)

Y2 - 5 July 0009 through 10 July 0009

ER -