IEEE Signal Processing Society Turkey Seminars

The IEEE Signal Processing Society organizes an event with 3 seminars on 27 August 2010 in collaboration with Özyeğin University.

During the event, the program will include the seminars entitled “Support Vector Machines: A Geometric Point of View” by Prof. Sergios Theodoridis from University of Athens,, “Information Theoretic Divergences for Recognition of Shapes and Structures” by Prof. Francisco Escolano from Alicante University, and  “Exploiting Context Information to Facilitate Semantics-Driven Nultimedia Applications” by Dr. Jiebo Luo from Eastman Kodak Company.

Date: 27 August 2010
Venue: Özyeğin University, Auditorium 1

Program:

11:00-12:00	“Support Vector Machines: A geometric point of view” Prof. Sergios Theodoridis, Fellow, IEEE IEEE Distinguished Lecturer, University of Athens
13:00-14:00	“Information Theoretic Divergences for Recognition of Shapes and Structures” Prof. Francisco Escolano, Universidad de Alicante
14:30-15:30	“Exploiting Context Information to Facilitate Semantics-Driven Multimedia Applications” Dr. Jiebo Luo, Fellow, IEEE, Eastman Kodak Company

Support Vector Machines: A Geometric Point of View

Prof. Sergios Theodoridis, University of Athens 
Support Vector Machines have been established as one of the major classification and regression tools for Pattern Recognition and Signal Analysis. Over the last decade, a number of theoretical arguments have been developed in order to justify their enhanced performance.

This talk will focus on the geometric approach and new results will be discussed concerning a) novel, necessary for our case, theorems concerning the structure and properties of the reduced convex hulls (RCH) and b) novel algorithms for computing the minimum distance between the resulting RCH´s. the SVM task.

Who is Prof. Sergios Theodorid?

Sergios Theodoridis is Professor of Signal Processing and Communications, Department of Informatics and Telecommunications, University of Athens. His research interests lie in the areas of adaptive algorithms and communications; machine learning and pattern recognition; and signal processing for audio processing and retrieval.

Dr. Theodoridis is currently Associate Editor, IEEE Transactions on Neural Networks, and IEEE Transactions on Circuits and Systems II, and Editorial Board Member, EURASIP Wireless Communications and Networking. He served as President of the European Association for Signal Processing (EURASIP) and is currently a member of the Board of Governors, IEEE CAS Society.

Dr. Theodoridis is co-author of four papers that have received best paper awards: IEEE Computational Intelligence Society Transactions on Neural Networks Outstanding Paper Award (2009); Best Student Paper Award of the IEEE Workshop on Multimedia and Signal Processing (2007); Best Paper Award of the Multimedia Metadata Applications (M3A) Workshop (2007); and Best Student Paper Award of the European Signal Processing Conference (EUSIPCO) (2005).

He was for four years, Member, Board of Directors of COSMOTE (the Greek mobile phone operating company). He is a Fellow of IET and a Fellow of IEEE.

Information Theoretic Divergences for Recognizing Shapes and
Structures

Prof. Francisco Escolano, Universidad de Alicante
Information-theoretic divergences (I-divergences) have been traditionally applied to image alignment problems (maximizing mutual information). Some of these measures have been proposed recently as optimization criteria for shape and points alignment up to 3D. However, the generalized use, proposal and evaluation of I-divergences for measuring both the similarity between shapes and the similarity between structural patterns (graphs) is an interesting challenge. The main problem for using such divergences is the need of estimating the probability density function (pdf) and some recent advances have been done recently for low dimensional data (shapes) due to the constraint imposed by the curse of dimensionality. However, modern entropy estimators show that it is possible to estimate I-divergences in multi- dimensional spaces through bypassing the pdf.

In this talk we will explore these bypass I-divergences and their applications to the shape and structural domains.

Who is Prof. Francisco Escolano?

Francisco Escolano obtained his Bachelors Degree in Computer Science at the Polytechnical University of Valencia (Spain) in 1992 and his Ph Degree in Computer Science at the University of Alicante in 1997. Since 1998, he has been an Associate Professor in the Computer Science and Artificial Intelligence Department at the University of Alicante. He has been post-doctoral fellow with Dr. Norberto M. Grzywacz at the Biomedical Engineering Department at the University of South California in Los Angeles, and he has also collaborated with Dr. Alan L. Yuille at the Smith-Kettlewell Eye Research Institute of San Francisco.

His research interests are focused on the development of efficient and reliable pattern recognition and computer vision algorithms for biomedical applications, bioinformatics, robotics and applications for the visually impaired. He is the head of the Robot Vision Group. He has recently published the book "Information Theory in Computer Vision and Pattern Recognitionand , since 2009, he is Vice-chair of the TC-2 IAPR Technical Committee.

Exploiting Context Information to Facilitate Semantics-Driven
Multimedia Applications

Dr. Jiebo Luo, Eastman Kodak Company
Lower cost devices and growing communication and internet infrastructure have led to an explosion in the creation, archival, distribution and consumption of multimedia (images, videos, music, and text/tags). Much of the active research has been focused on semantic multimedia understanding; with a growing emphasis on finding ways to overcome the well known semantic gap and intent gap. Context is a powerful cue in the human recognition process where the humans make extensive use of the environmental knowledge to facilitate object and scene recognition. Likewise, context can be used to improve the performance of automated systems.

We present an overview on exploiting a broad array of context information, including scene context, geo-location context, and social context, to facilitate semantics- driven multimedia applications.

Who is Dr. Jiebo Luo?

Jiebo Luo is a Senior Principal Scientist with the Kodak Research Laboratories in Rochester, NY. He received a B.S. degree and an M.S. degree in Electrical Engineering from the University of Science and Technology of China (USTC) in 1989 and 1992, respectively, and a Ph.D. degree in Electrical Engineering from the University of Rochester in 1995. His research interests include signal and image processing, pattern recognition, computer vision, and the related multi-disciplines such as multimedia data mining, biomedical informatics, computational photography, human-computer interaction, and ubiquitous computing.

Dr. Luo has authored over 150 papers and holds over 50 granted US patents. He was the recipient of the Best Poster Paper Award at the 2008 ACM International Conference on Content-based Image and Video Retrieval. He is the Editor- in-Chief of the Journal of Multimedia (Academy Publisher). He has also served on the editorial boards of the IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), the IEEE Transactions on Multimedia (TMM), the IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), Pattern Recognition (PR), Machine Vision Applications (MVA), and Journal of Electronic Imaging (JEI). He is a Kodak Distinguished Inventor, a winner of the 2004 Eastman Innovation Award, a Fellow of SPIE, a Fellow of IAPR, and a Fellow of IEEE.