Skip to Main content Skip to Navigation
Conference papers

Multimodal Prior Appearance Models based on Regional Clustering of Intensity Profiles

François Chung 1 Hervé Delingette 1, * 
* Corresponding author
1 ASCLEPIOS - Analysis and Simulation of Biomedical Images
CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : Model-based image segmentation requires prior information about the appearance of a structure in the image. Instead of relying on Principal Component Analysis such as in Statistical Appearance Models, we propose a method based on a regional clustering of intensity profiles that does not rely on an accurate pointwise registration. Our method is built upon the Expectation-Maximization algorithm with regularized covariance matrices and includes spatial regularization. The number of appearance regions is determined by a novel model order selection criterion. The prior is described on a reference mesh where each vertex has a probability to belong to several intensity profile classes.
Complete list of metadata

Cited literature [11 references]  Display  Hide  Download
Contributor : Project-Team Asclepios Connect in order to contact the contributor
Submitted on : Monday, July 8, 2013 - 5:06:41 PM
Last modification on : Saturday, June 25, 2022 - 11:06:37 PM
Long-term archiving on: : Wednesday, October 9, 2013 - 2:50:12 AM


Files produced by the author(s)




François Chung, Hervé Delingette. Multimodal Prior Appearance Models based on Regional Clustering of Intensity Profiles. Medical Image Computing and Computer-Assisted Intervention (MICCAI'09), 2009, London, United Kingdom. pp.1051--1058, ⟨10.1007/978-3-642-04271-3_127⟩. ⟨inria-00616132⟩



Record views


Files downloads