Parameterisation invariant statistical shape models

Johan Karlsson; Anders Ericsson; Karl Åström

doi:10.1109/ICPR.2004.1333696

Parameterisation invariant statistical shape models

Johan Karlsson, Anders Ericsson, Karl Åström

Matematik LTH

Forskningsoutput: Kapitel i bok/rapport/Conference proceeding › Konferenspaper i proceeding › Peer review

Sammanfattning

In this paper novel theory to automate shape modelling is described. The main idea is to develop a theory that is intrinsically defined for curves, as opposed to a finite sample of points along the curves. The major problem here is to define shape variation in a way that is invariant to curve parametrisations. Instead of representing continuous curves using landmarks, the problem is treated analytically and numerical approximations are introduced at the latest stage. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to global reparametrisations. An algorithm for implementing the ideas is proposed and compared to a state of the an algorithm for automatic shape modelling. The problems with instability in earlier formulations are solved and the resulting models are of higher quality

Originalspråk	engelska
Titel på värdpublikation	Proceedings of the 17th International Conference on Pattern Recognition
Förlag	IEEE - Institute of Electrical and Electronics Engineers Inc.
Sidor	23-26
ISBN (tryckt)	0-7695-2128-2
DOI	https://doi.org/10.1109/ICPR.2004.1333696
Status	Published - 2004
Evenemang	17th International Conference on Pattern Recognition, 2004 - Cambridge, England, Cambridge, Storbritannien Varaktighet: 2004 aug. 23 → 2004 aug. 26

Konferens

Konferens	17th International Conference on Pattern Recognition, 2004
Land/Territorium	Storbritannien
Ort	Cambridge
Period	2004/08/23 → 2004/08/26

Ämnesklassifikation (UKÄ)

Matematik

Tillgång till dokument

10.1109/ICPR.2004.1333696

Citera det här

@inproceedings{6996ca54c0a146738de4246cfb33d953,

title = "Parameterisation invariant statistical shape models",

abstract = "In this paper novel theory to automate shape modelling is described. The main idea is to develop a theory that is intrinsically defined for curves, as opposed to a finite sample of points along the curves. The major problem here is to define shape variation in a way that is invariant to curve parametrisations. Instead of representing continuous curves using landmarks, the problem is treated analytically and numerical approximations are introduced at the latest stage. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to global reparametrisations. An algorithm for implementing the ideas is proposed and compared to a state of the an algorithm for automatic shape modelling. The problems with instability in earlier formulations are solved and the resulting models are of higher quality",

keywords = "automatic shape modelling, invariant statistical shape models, covariance matrix, curve parametrisations",

author = "Johan Karlsson and Anders Ericsson and Karl {\AA}str{\"o}m",

year = "2004",

doi = "10.1109/ICPR.2004.1333696",

language = "English",

isbn = "0-7695-2128-2",

pages = "23--26",

booktitle = "Proceedings of the 17th International Conference on Pattern Recognition",

publisher = "IEEE - Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "17th International Conference on Pattern Recognition, 2004 ; Conference date: 23-08-2004 Through 26-08-2004",

}

Karlsson, J, Ericsson, A & Åström, K 2004, Parameterisation invariant statistical shape models. i Proceedings of the 17th International Conference on Pattern Recognition. IEEE - Institute of Electrical and Electronics Engineers Inc., s. 23-26, 17th International Conference on Pattern Recognition, 2004, Cambridge, Storbritannien, 2004/08/23. https://doi.org/10.1109/ICPR.2004.1333696

TY - GEN

T1 - Parameterisation invariant statistical shape models

AU - Karlsson, Johan

AU - Ericsson, Anders

AU - Åström, Karl

PY - 2004

Y1 - 2004

N2 - In this paper novel theory to automate shape modelling is described. The main idea is to develop a theory that is intrinsically defined for curves, as opposed to a finite sample of points along the curves. The major problem here is to define shape variation in a way that is invariant to curve parametrisations. Instead of representing continuous curves using landmarks, the problem is treated analytically and numerical approximations are introduced at the latest stage. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to global reparametrisations. An algorithm for implementing the ideas is proposed and compared to a state of the an algorithm for automatic shape modelling. The problems with instability in earlier formulations are solved and the resulting models are of higher quality

AB - In this paper novel theory to automate shape modelling is described. The main idea is to develop a theory that is intrinsically defined for curves, as opposed to a finite sample of points along the curves. The major problem here is to define shape variation in a way that is invariant to curve parametrisations. Instead of representing continuous curves using landmarks, the problem is treated analytically and numerical approximations are introduced at the latest stage. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to global reparametrisations. An algorithm for implementing the ideas is proposed and compared to a state of the an algorithm for automatic shape modelling. The problems with instability in earlier formulations are solved and the resulting models are of higher quality

KW - automatic shape modelling

KW - invariant statistical shape models

KW - covariance matrix

KW - curve parametrisations

U2 - 10.1109/ICPR.2004.1333696

DO - 10.1109/ICPR.2004.1333696

M3 - Paper in conference proceeding

SN - 0-7695-2128-2

SP - 23

EP - 26

BT - Proceedings of the 17th International Conference on Pattern Recognition

PB - IEEE - Institute of Electrical and Electronics Engineers Inc.

T2 - 17th International Conference on Pattern Recognition, 2004

Y2 - 23 August 2004 through 26 August 2004

ER -

Parameterisation invariant statistical shape models

Sammanfattning

Konferens

Ämnesklassifikation (UKÄ)

Tillgång till dokument

Fingeravtryck

Citera det här