Tutorials
The role of the tutorials is to provide a platform for a more intensive scientific exchange amongst researchers interested in a particular topic and as a meeting point for the community. Tutorials complement the depth-oriented technical sessions by providing participants with broad overviews of emerging fields. A tutorial can be scheduled for 1.5 or 3 hours.
Low Rank and Sparse Representations in Large Scale Image Processing
Instructor
|
Tien Dai Bui
Computer Science and Software Engineering, Concordia University
Canada
|
|
|
|
Brief Bio
Professor Tien D. Bui has been a member of the Centre for Pattern Recognition and Machine Intelligence and a full professor in the Department of Computer Science and Software Engineering at Concordia University in Montreal, Canada for more than 30 years. He taught at McGill University from 1971 to 1974. Since 1974 he has held various positions at Concordia, Chair of the Department in 1985-90, Associate Vice-Rector Research in 1992-96, and member of Concordia University Senate and its Board of Governors. He held various positions on many Boards of Directors in Canada. Dr. Bui is currently an Associate Editor of the journal Signal Processing (EURASIP), the Int. Journal of Wavelets, Multi-resolution and Information Processing, and the Journal of Wavelet Applications and Theory. He served on many organizing and program committees of international conferences and on many national grant selection committees. Professor Bui has received many research grants and published over 250 papers. He received many Best Paper Awards with his students including the 9th IWFHR award in Tokyo, 2006; the EURASIP award in 2010 for a paper published in EURASIP journals over the five-year period 2005-2010; the International Joint Conference on Biometrics (IJBM) award in Washington DC, 2011. He is co-author of the book Computer Transformation of Digital Images and Patterns (World Scientific, 1989). He has been a visiting professor at Berkeley, Paris V, and the Istituto per le Applicazioni del Calcolo (Rome). His current research interests include image processing, computer vision, biometrics, machine learning, pattern recognition, and document analysis. He is a Life Senior Member of the IEEE.
|
Abstract
In the last few years there has been a lot of attention to the techniques of sparse representation and low-rank approximation; and these techniques have shown important applications in many areas in image and video processing, signal analysis, computer vision and machine learning. These techniques can be used successfully in unsupervised and dictionary learning to uncover high-order relations in the data and to train deep neural networks. However it is only the beginning that these techniques are used in digital document processing. In this tutorial I will discuss the latest developments and show the effectiveness of these techniques in large scale image processing systems.
Keywords
Low-rank Approximation, Sparse Representation, Image and Video Processing, Computer Vision.
Aims and Learning Objectives
The aim is to learn the latest techniques in low-rank and sparse representations and their applications.
Target Audience
Graduate students, researchers, engineers.
Prerequisite Knowledge of Audience
Some background in advanced linear algebra, general math and probability theory.
Detailed Outline
1)Some problems in low-rank approximation and sparse representation.
2)Non-convex and Lp-norm optimization problems
3)Lp-norm solution of Robust PCA and Robust SVD
4)Applications of the matrix decomposition problems