Course Description

Adaptive Image Representations

The course will take place on Monday March 21st, 2011, in Louvain-la-Neuve (Croix du Sud 4-5, room  B059).

Abstract
Improving the modeling of natural images is important to go beyond the state of the art for many image processing tasks such as compression, denoising, inverse problems and texture synthesis. Natural images are composed of intricate patterns such as regular areas, edges, junctions, oriented oscillations and textures. Processing efficiently such a wide range of regularities requires methods that are adaptive to the geometry of the image. This adaptivity can be achieved using sparse representations in a redundant dictionary. The geometric adaptivity is important to search for efficient representations in a structured dictionary. Another way to capture this geometry is through non-local interactions between patches in the image. The resulting non-local energies can be used to perform an adaptive image restoration.

This course will review these emerging technics and shows the interplay between sparse and non-local regularizations.
In particular, Gabriel Peyré will introduce the following concepts:

 * sparse image representations and adaptive bases (such as Bandlets),
 * the sparsity principle in image restoration and image processing applications (denoising, deconvolution, inverse problems, compressed sensing),
 * non-local processing of images (NL-means, spectral basis, Total Variation on graph).

The course will be split in two parts of 1h30 duration each:
    from 11h to 12h30 and from 14h to 15h30, Monday March 21st, 2011 

Short biography of G. Peyre :
Gabriel Peyré graduated from Ecole Normale Supérieure de Cachan, France, in 2003 and received his Ph.D in applied mathematics from école Polytechnique, Paris, France, in 2005. His research interests include geometrical representations for image processing with applications in computer and biological vision.

Gabriel Peyré worked with Stéphane Mallat on image compression with bandlets during his Ph.D. Since 2006, he has been a Researcher at the Centre Nationale de Recherche Scienti?que (CNRS), working in Ceremade, University Paris-Dauphine. Since 2005 Gabriel Peyré has co-authored 15 papers in international journals, 30 conference proceedings in top vision and image processing conferences, and two books. He was the main contributor of the 3rd edition of the book "Wavelet tour of signal processing" of S. Mallat.

He is the creator of the "Numerical tour of signal processing", a popular online repository of Matlab/Scilab ressources to teach modern signal and image processing (http://www.numerical-tours.com).

Registration :
Free but *requested*. Please consult:
http://www.tele.ucl.ac.be/musics/courses.php


Venue :
http://www.uclouvain.be/en-acces-lln.html

External visitors are welcome to park on the Redime parking.
http://www.uclouvain.be/9913.html

Exact course location : Croix du Sud 4-5, room B059.  See map on square E8 
http://uclouvain.be/cps/ucl/doc/adpi/documents/PLAN_2007recto.pdf