Seminars/Colloquia/Invited Talks

Seminars

Co-hosted by the Applied Math and Statistics Department

Alain Trouve

Can we "really" learn Deformable Models?

PLACE:	Whitehead 304
EVENT:	CIS Seminar
DATE:	October 20, 2005
TIME:	4:00 - 5:00

Abstract

The link between bayesian and variational approach in image processing is common knowledge in particular in the context of deformable models. However, true generative models and consistent estimation procedures are usually not available and the main stream is the computation of statistics, mainly based on PCA analysis. We will talk about a recent study, done with S. Allassonnière and Y. Amit, showing that thanks to a careful statistical modeling of deformable structures, we can derive an effective and consistent estimation algorithm for the various parameters (geometric and photometric) appearing in the model.