[visionlist] Registration open for Deep Hierarchies in Vision Workshop (in conj. with CogSys2012)

Sinan KALKAN sinankalkan at googlemail.com
Tue Jan 17 07:58:36 GMT 2012

Please circulate freely.

Our Workshop has received excellent submissions and has been accepted for a
generous support
from the PASCAL2 Network of Excellence (http://pascallin2.ecs.soton.ac.uk/).
The registration
to the workshop (free of charge) and the program are available online at:

*Workshop on Deep Hierarchies in Vision:*
(in conjunction with CogSys2012 - 5th Int. Conf. on Cognitive Systems)
Webpage: http://www.kovan.ceng.metu.edu.tr/~sinan/DHV/

=> February 21, 2012; Vienna-Austria <=

Organizers: Ales Leonardis, Norbert Krueger, Richard Bowden, Sinan Kalkan,
Nicolas Pugeault, Frank Guerin
Webpage for the 5th Int. Conf. on Cognitive Systems:


Processing in the brain in general and visual processing in particular is
organized in a hierarchical fashion, from simple localized features to
complex, large scale features. The visual system consists of a hierarchy,
in which neurons in early visual areas extract simple image features
(orientation, motion, disparity) over a small local region of visual space,
which are then transmitted to neurons in higher visual areas responding to
more complex features (e.g. shape) over a larger region of visual space.
Hierarchical representations can derive and organize features at multiple
levels. Hierarchical representations with many levels are called 'deep
hierarchies'. They build on top of each other by exploiting the
shareability of features among more complex compositions or objects
themselves. Sharing features, on the one hand, means sharing common
computations, which brings about (highly desirable) computational
efficiency. On the other hand, reusing the commonalities between objects’
models places their representations in relation to other objects, thus
leading to high generalization capabilities and lower storage demands.

However, although all neurophysiologic evidence suggests that in the human
visual system quite a number of levels are realized, it has turned out that
the design and/or learning of such deep hierarchical systems is a very
difficult task. Most existing computer vision systems are 'flat' (e.g.,
having rather simple features - such as SIFT - as input and then applying
some kind of SVM learning) and hence cannot make use of the advantages
connected to deep hierarchies. Here in particular the generalization
capabilities are crucial for any form of cognitive intelligence. As a
consequence, we see the issue of establishing deep hierarchies as one major
challenge for the establishment of truly cognitive systems.

The aim of the workshop is to bring together researchers from vision,
robotics, machine learning, artificial intelligence, and neurophysiology to
discuss existing obstacles in the design of deep hierarchies, possible
solutions as well as perspectives for deep hierarchies in vision and

*Invited speakers:*

"Learning Deep Architectures: Some Successes and Problems"
Christian Igel
The Image Group,
University of Copenhagen, Denmark

"Spatial Distributions, Hierarchical Models and Visual Inference"
Justus Piater
Department of Electrical Engineering and Computer Science
University of Innsbruck, Austria

"Hierarchical Processing with Slow Feature Analysis"
Laurenz Wiskott
Institut für Neuroinformatik
Ruhr-Universität Bochum, Germany

"The Neural Representation of the Three-Dimensional Shape in the Primate
Peter Janssen
Division of Neurophysiology
Katholieke Universiteit Leuven, Belgium


Online registration is necessary for the arrangement of the resources at
following website:

Registration to the workshop is free thanks to the PASCAL2 Network of

*Program Co-chairs:*

Laurenz Wiskott
Institut für Neuroinformatik
Ruhr-Universität Bochum, Germany

Peter Janssen
Division of Neurophysiology
Katholieke Universiteit Leuven, Belgium


Ales Leonardis, ales.leonardis at fri.uni-lj.si
Norbert Krueger, norbert at mmmi.sdu.dk
Sinan Kalkan, skalkan at ceng.metu.edu.tr

Sinan KALKAN, Asst. Prof.
Dept. of Computer Engineering
Middle East Technical University
Web: http://kovan.ceng.metu.edu.tr/~sinan
Tel:  + 90 - 312 - 210 5547  / + 90 - 312 - 210 7371
Fax: + 90 - 312 - 210 5544
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist/attachments/20120117/1bf74853/attachment.htm>

More information about the visionlist mailing list