###########################################################################
 
                        SECOND CALL FOR PAPERS
 
                        ---------------------
                        | EXTENDED DEALINES |
                        ---------------------
 
                        Special Session on
 
        ADVANCED CONCEPTS FOR INTELLIGENT VISION SYSTEMS (ACIVS'99)
 
 
                        held in conjunction with
 
        The  11-th  International Conference on Systems Research,
                        Informatics and Cybernetics
 
 
                        2 - 7 August 1999
                        Baden-Baden (Germany)
 
 
###########################################################################
 
                        SECOND CALL FOR PAPERS
 
 
 
* GENERAL SCOPE
   
     Image  Processing gathers a wide field  of approaches  which  aim at
   making systems  as autonomous as  possible. Hundreds  of sophisticated
   techniques do exist in the toolbox of  today's engineers which help to
   cope   with a  number of  industrial  and  scientific applications  in
   constant growth.
     
     However,  there is currently  no  mature common framework for  using
   them and understanding the mechanisms at work from  the early stage of
   low level  Image Processing to the highest  level of Vision.  Whatever
   the vision      paradigm  at work  is    (from  data-driven  vision to
   goal-directed vision), algorithms   have to  be tunable or   adaptive,
   their performances  have to be  assessed and  one must exhibit control
   laws of their parameters with respect to their result. It is the price
   to   be paid  for plugging  software  boxes  into bigger  and coherent
   complex systems.
   
     This special  session  is mainly concerned with  providing (partial)
   answers   to this particular scientific   issue; of course, any papers
   presenting novel Image Processing techniques are welcome.
 
 
* TOPICS include but are not limited to
 
   1) THEORETICAL TOOLS FOR A UNIFIED FRAMEWORK: in order to embed the
   problem within the  correct theoretical background, the  first track
   deals with any aspect of  the relationship between the digital image
   and the symbolic and semantic space.
   
   - measures of information, relationship between numerical and symbolic 
     information
   - multiresolution theory
   - scale-space theory
   - variational methods
   - fractals and multifractals
   - formal languages theory for pattern recognition
   - biologically-inspired vision
 
   2) OPTIMAL  AND ADAPTIVE ALGORITHMS:  adaptive algorithms  have been
   designed for taking into account   the large variability of   online
   conditions.  Such   algorithms  yield results whose  resolution  and
   accuracy are  a function of  the data  itself.  The impact  of local
   adaptivity on  the design of systems   must be considered throughout
   topics like:
 
   - adaptive filtering and image restauration
   - adaptive segmentation
   - model-based algorithms
   - semantic image compression
   - tools for adaptive algorithms
 
   3) ASSESSMENT OF IMAGE PROCESSING   ALGORITHMS: whatever the control
   law of an  algorithm may be, it must  on the one  hand operate on an
   efficient  algorithm  and on the other   hand take into account some
   quality criterion of   the  computed result.  Valid   topics include
   scientific   studies (results    comparisons,  performance metrics),
   reviews of available  techniques (benchmarks and frameworks) as well
   as industrial applications.
 
   - objective metrics
   - perceptually based distortion metrics
   - video quality evaluation
   - subjective ratings of compression algorithms
   - professional applications (medical, military, etc.)
   - empirical methods for assessing IP algorithms
   - software tools and databases for performance evaluation
 
   4) FROM TUNING TO CONTROLLING VISION SYSTEMS: once vision components
   have  been  optimized, adapted  and assessed  and  that the  complex
   system running them has  been  characterized, this latter must  know
   how to  control its "organs"to  achieve its task. Obviously, such an
   issue can  be adressed  today as a   whole and thus,   the following
   topics focus on the tools needed for setting up the problem.
 
   - model selection and validation
   - fuzzy modelling for pattern recognition and vision applications
   - nonlinear system identification  
   - learning and nonlinear optimization   
   - multilayer perceptrons             
   - simulated annealing
   - recurrent networks                  
   - genetic algorithms
 
 
* IMPORTANT DATES
 
    June 23, 1999 (Wednesday):  Full papers due (4 / 8 pages) 
       July 12, 1999 (Monday):  Notification of acceptance
                July 27, 1999:  Camera-Ready papers
           August 2 - 7, 1999:  ACIVS Symposium
 
 
* ELECTRONIC  ABSTRACT AND PAPER SUBMISSION
 
     The maximum length for full  papers is 8 pages. Submissions should
   preferably  be in LaTex format (12pt,  with standard LaTex style and
   figures as encapsulated    PS) but  may  also be  in another  format
   provided it has been approved by the co-chairmen.  Submissions  must
   be sent by e-mail to either:
 
   Jacques.Blanc-Talon@etca.fr
   Dan.Popescu@cmis.csiro.au
 
   More information on the conference can be found at: 
   http://www.iias.edu
 
 
* MANUSCRIPT SUBMISSION - REVIEW PROCESS 
 
        All    submissions  will  be  reviewed    by   members  of  the
   International programme Committee  but additional reviewers will  be
   consulted if needed.  The average review  time is about three weeks.
   Authors  of accepted papers will be  informed in time  (by e-mail if
   available)  of  the    required   format  for   camera-ready   paper
   submissions.    In order for   reviewers to  be  able  to assess the
   submissions,  papers have  to provide sufficient  material about the
   background to  the problem, the  novelty of the obtained results and
   the    results achieved, the  conclusions   drawn   and some  recent
   references.  Up to three keywords should be supplied.  All submitted
   papers  have to  be   original, unpublished and   not submitted  for
   publication elsewhere.
      
      
* PROCEEDINGS 
 
     All accepted   papers   will  be   published in   the   Conference
   Proceedings, but due  to  the short time  between  this CFP and  the
   conference,   the   proceedings  may   not  be    available  at  the
   conference. In any case, they will be sent to the authors as soon as
   possible. 
 
* PROGRAM COMMITTEE
 
 Pr Hamid R. Arabnia, University of Georgia, Athens, GA, USA. 
 Dr Mark Berman, CSIRO, Macquarie University campus, NSW, Australia.
 Dr Jacques Blanc-Talon, CTA, Arcueil, France.
 Pr Philippe Bolon, LAMII, University of Savoie, France.
 Dr Don Bone, CSIRO, Canberra, ACT, Australia.
 Pr Horst Bunke, IAM, Bern, Switzerland.
 Pr Touradj Ebrahimi, EPFL, Lausanne, Switzerland.
 Dr Georgy Gimel'farb, University of Auckland,  New Zealand.
 Pr Hideo Kuroda, Nagasaki University, Japan.
 Dr Craig A. Lindley, CSIRO, Macquarie University campus, NSW, Australia.
 Pr Marcin Paprzycki,  University  of   Southern   Mississippi, MS, USA.
 Pr Sylvie Philip, ENSEA, Cergy, France.
 Pr Wilfried Philips,  Vakgroep TELIN, Gent, Belgium.
 Dr Dan Popescu, CSIRO, Canberra, Australia.
 Pr Georges Stamon, IARFA, Paris VI University, France.
 
* SYMPOSIUM CO-CHAIRS
   
   Dr Jacques Blanc-Talon
   Geographie-Imagerie-Perception
   CTA/GIP
   16 bis, Avenue Prieur de la cote d'or,
   94114, Arcueil, FRANCE
   Jacques.Blanc-Talon@etca.fr
 
   Dr Dan Popescu 
   Division of Mathematical and Information Science
   CMIS/CSIRO
   GPO Box 664
   Canberra, ACT  2601  AUSTRALIA
   Dan.Popescu@cmis.CSIRO.AU
 
 
* GENERAL CONFERENCE CHAIR
 
   Prof. George E. Lasker
   I.I.A.S. & Department of Computer Science
   University of Windsor
   Windsor, Ontario, CANADA