Course Description

Sparse Representation of Signals: Theory, Algorithms and Applications

By Prof. Pierre Vandergheynst, Signal Processing Laboratory (LTS2), EPFL, Lausanne, Switzerland

The course will take place on Thursday May 6 and Friday May 7, 2010 in Louvain-la-Neuve.

Abstract
Sparsity models are nowadays very popular in many modern signal/image processing topics. Their reputation comes initially from works in non-linear approximation theory where many functional spaces have been represented thanks to sparse decompositions in wavelet basis.
These results were extended later to other signal representations (time-frequency, geometrical image models, ...) and soon, the sparsity concept motivated by itself a huge set of more fundamental questions: How to identify a sparse signal representation? How stable this decomposition is? Is it Unique? In parallel to these theoretical questions, many applications have been successfully solved by the sparsity paradigm (e.g. signal denoising, image deconvolution, general inverse problems, dictionary learning, ...) and these have fed back numerous advances in numerical methods computing sparse representations.

In this course, we will come back on the long travel that has been realized since the first sparsity model formulation. We will overview various applications of sparsity in signal/image processing, merging fundamental and practical results. Recent and promising geometric generalizations of sparse signal representation will be discussed, together with their connections in the resolution of particular inverse problems as in Compressed Sensing and Data Regularization on Graph.     
 

Short Biography
Pierre Vandergheynst received the M.S. degree in physics and the Ph.D. degree in mathematical physics from the Université catholique de Louvain, Louvain-la-Neuve, Belgium, in 1995 and 1998, respectively. From 1998 to 2001, he was a Postdoctoral Researcher with the Signal Processing Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. He was Assistant Professor at EPFL (2002-2007), where he is now an Associate Professor. His research focuses on harmonic analysis, sparse approximations and mathematical image processing with applications to higher dimensional, complex data processing. He was co-Editor-in-Chief of Signal Processing (2002-2006) and is Associate Editor of the IEEE Transactions on Signal Processing (2007-present). He has been on the Technical Committee of various conferences and was Co-General Chairman of the EUSIPCO 2008 conference. Pierre Vandergheynst is the author or co-author of more than 50 journal papers, one monograph and several book chapters. He’s a senior member of the IEEE, a laureate of the Apple ARTS award and holds seven patents.

The doctoral course will be accompanied by other talks on related topics by
 - Dr Laurent Jacques (TELE/UCL),
 - Prof. Christophe De Vleeschouwer (TELE/UCL),
 - Thibault Hellepute (INGI/UCL)
 - Alexandre Alahi (LTS2/EPFL, Switzerland)
 - Emmanuel Foumouo ( IMCN,UCL)


Schedule
Thursday May 6, 2010:
09h30 - 11h00: Lecture by P. Vandergheynst
11h00 - 11h15: Break
11h15 - 12h45: "Sparsity and Greedy Methods in Image and Video Processing" by C. De Vleeschouwer (TELE/UCL)

13h30 - 15h00: Lecture by P. Vandergheynst
15h00 - 15h15: Break
15h15 - 15h45: "Sparsity and Automatic Pedestrian Detection", by A. Alahi (LTS2/EPFL)
15h45 - 16h15: "Sparsity in Optics", E. Foumouo ( IMCN,UCL)

Friday May 7, 2010:
 09h30 - 11h00: Lecture by P. Vandergheynst
11h00 - 11h15: Break
11h15 - 12h45: Lecture by P. Vandergheynst

14h00 - 15h00: "Compressed Sensing and Applications" by L. Jacques (TELE/UCL)
15h00 - 15h30: "Sparsity in Gene Expression Analysis", by T. Helleputte  (INGI/UCL)

Location : Aud. Barb 11

Related event
We enjoy of this occasion to mention that a related (CIL) doctoral course on "Sparse methods for machine learning: theory and algorithms"  will be given by Francis Bach (INRIA -- Ecole Normale Supérieure, Paris, France), in Louvain-la-Neuve.
The course will take place on Wednesday May 5, 2010 from 10.30 to 13.00 and 14.30 to 17.00. 

More details will be provided soon by the organizers on http://www.uclouvain.be/doctoralschool-cil

Registration
Registration is 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