Newsletter 33


- Some comments about IWGR'95
- Report about ICDAR'95
- CFP: DAS'96
- CFP: Performance Characterisation Workshop
- CFP: CVPR'96
- CFP: ICMI'96
- CFP: BMVC'96
- Provisional program: BMVA Document image processing and multimedia
- Advance program & registration info: ICSC'95
- Schedule: AAAI'95 Fall symposim

Dear TC10 members and friends,

Here comes again some info from TC10. First, I invite you all to
access additional information on my WWW pages. My general WWW page is
at URL http://www.loria.fr/~tombre/
Then you have the tc10 page at http://www.loria.fr/~tombre/tc10.html,
where you can also find the archives of this mailing list.
I also have a list of pointers to additional information at URL
http://www.loria.fr/~tombre/vision.html, where you can also find the
archives of CFPs and advance programs announced here, and some others
which I don't announce here because they are too far from the scope of
interest of this mailing list. Hence, I have also shortened some of
the announcements in this list, as the full copy is available on the web.

I have received a report from two colleagues with some remarks about
the scientific content of IWGR. They raise several good questions,
which we should take into account to do it still better next
time. Thank you for these comments. Of course, everybody is free to
comment on these comments again; it would be to the benefit of the
whole mailing list to have some discussions going on. You can send
your comments directly to the whole list by sending to the address
This email address is being protected from spambots. You need JavaScript enabled to view it. , or you can send it to me for inclusion in the next TC10


Remarks on the IWGR
Here are some comments about the 1st IWGR:

First, congratulations for the general organization of the workshop,
especially the panels which dealt with ideal situations to discuss and
to combat the hot issues or serious problems. In our opinion, during
the workshop the panel discussions were the most valuable ones.

Secondly, we think that the principal themes were divided into 3
different categories as follows:
1. Basic techniques and new results
2. Some special real applications
3. Performance measures and evaluation

Although many interesting papers were presented, we found that more
profound articles were sometimes missing. Therefore, we could have
taken more times for discussions. The missing factors are summarized
as follows:

For the first theme; the review on the posed problems, existing
techniques and their performances were lacking. If the commercialized
tools and the research activities were mentioned, it would have been
very interesting. One of us came to the workshop looking for specific
methods for solving the line extraction problem. There were actually
very few articles on this topic. Adbel's article on preprinted forms,
which seemed to be similar to Chhabra's article, was not presented in
the correct session. If the papers had presented together, the line
extraction problem on the printed forms could have been given a more
precise comparison during the discussion.

The second theme was well organized. Dov's presentation gave a good
overview of this domain. An excellent survey on the real problem and the
principal results were demonstrated. What was maybe lacking a bit was:
- the databases: names, standards, etc.
- the performance comparison between the systems: maybe this
stems from the fact that eveyrbody works on his/her own
application and uses theregore ad hoc databases
- maps processing could have been given more time
- Vectorization techniques and raster processing techniques
should have been added, with respect to the depth and strength
of the basic ideas. However, some interesting real
application papers (ex. Kwon and den Hartog) were presented.
- 3D reconstruction was a very interesting topic, but a more
general overview of its benefit and its duality with CAD was missing.
- Models: few things were discussed on this matter, except Dov's
work on object modeling.
- The interpretation stategies could have been given more
time. The difference between top-down and bottom-up methods
was introduced by Jurgen, but many other approaches were also
possible. Knowledge-based systems can use the best of both approaches.
- We missed references and links to similar work in hanwriting

With respect to the third theme, performance measurement is a very
interesting subject, but its presentation was not adequate
enough. We were expecting more concrete results and ideas.
For instance, Abe's paper on the comparison of several existing
methods was very interesting but would have been still better if
provided with a complete benchmark and a summary of the most important
parameters. In each algorithm, the most important factor was the
threshold. How should we calculate and adapt the threshold? This was a
good discussion topic during the workshop.

We have highlighted some shortcomings and critics, but we think that
the IWGR was very well organized and that the contents of the papers
had good quality. We strongly recommand a second workshop for the further
development of the graphics recognition topic.

Abdel Belaid, CRIN/CNRS & INRIA-Lorraine, France
Young-Bin Kwon, Chungang University, Seoul, Korea

Prof. Nakano also sent me a short report about ICDAR'95:
>From: NAKANO Yasuaki <This email address is being protected from spambots. You need JavaScript enabled to view it. >
>Subject: Report of ICDAR'95

Third International Conference on Document
Analysis and Recognition
Montreal, Canada, 14-16 August 1995

The IAPR Conference ICDAR'95 was held on 14-16 August 1995 at Montreal
Canada, hosted by General Chair, Prof. C. Y. Suen, Concordia University.

Number of participants from 25 countries ran up to 437, and exceeded those
of two predecessors, ICDAR'91 held at Saint-Malo, France and ICDAR'93 at
Tsukuba, Japan. It should be noted that more than a half participants came
from outside of North America. ICDAR'95 was really a worldwide forum for
document analysis/recognition. In a rough count, a half members came from
North and South America, and a quarter from Asia and Europe, respectively.
Besides, it should also be noted that about 180 participants were from
companies, so ICDAR'95 served as a junction point between academic and
industrial societies.

Dr. P. E. Baier from Universitaet Mannheim in Germany gave a Keynote
Speech entitled "Image Processing of Forensic Documents" which was an
overview of the potential application of the document analysis/recognition

>From about 330 papers submitted, 12 long papers, 169 short papers and
96 posters were accepted. The quality of the papers is high enough and
participants discussed enthusiastically in every session.

The topics of oral sessions were as follows:
Document (Structure and) Analysis, Model-based Document Analysis,
Layout Analysis, Skew Detection and Processing, Image Processing
Techniques, Segmentation, Word (and Character) Recognition,
Handwritten Character Recognition, On-line Recognition, Signature
Verification, Drawing and Map Recognition, Map Interpretation,
Symbol Recognition, Theoretical Approaches and Music Recognition,
Modelling Methods, Learning, Neural Networks (and Systems),
Postprocessing, Application Systems, Database and Document Retrieval,
Storage and Retrieval.

These topics were discussed in 36 oral sessions and poster sessions.
Oral sessions were arranged in four parallel tracks because of the
limited schedule.

As my impression, character recognition techniques are now discussed in
the word level, including cursive script, rather than in the level of
already separated characters. Separation of touching characters and
extraction of characters from complex backgrounds are also the current
problems. These problems were, of course, main themes in many sessions.
Techniques of document analysis and map interpretation, corporating with
character/symbol recognition as well as theoretical considerations and
applications were also important aspects in the conference.

Along with Poster Sessions, demonstrations related to the papers
were held using PC's or Workstations, so the audience could recognize
and evaluate the achievements more quantitatively and qualitatively.

>From the papers given in Poster Sessions, the following three papers
were awarded as the Best Posters:
# Myriam Cote, Eric Lecolinet, Mohamed Cheriet and C. Y. Suen,
"Building a Perception Based Model for Reading Cursive Script."
# Oivind Due Trier, Torfinn Taxt and Anil K. Jain,
"Data Capture from Maps Based on Gray Scale Topographic Analysis."
# Donald Sylwester and Sharad Seth,
"A Trainable, Single Pass Algorithm for Column Segmentation."

As seen from the explanation above, ICDAR is the worldwide and largest
forum for introduction and exchange of academic and industrial ideas
on these fields. There may be no doubt in that the researches and
applications in these fields will proliferate much more in coming years.

Finally, representing all participants, I want to appreciate the labor
and hospitality of the Local Arrangement Committee at Montreal.

The next conference, ICDAR'97, will be held on 18-20 August 1997 at Ulm,
Germany, hosted by General Chair Prof. J. Schuermann, Daimler-Benz.
ICDAR'99 will be held on 4-6 October 1999 at Bangalore, India. Prof.
Srihari, SUNY Buffalo, will serve as the General Chair.

Reported by
Yasuaki Nakano
Shinshu University
Publicity Co-Chair, ICDAR'95
This email address is being protected from spambots. You need JavaScript enabled to view it.

Here is an announcement I received recently:
>From: Thomas M. J. Fruchterman <This email address is being protected from spambots. You need JavaScript enabled to view it. >
>Subject: DAFS 0.9.2 -- OCR and Document Understanding Format

DAFSLib 0.9.2
A Format for OCR and Document Decomposition

RAF Technology would like to announce the latest release of the
Document Attibute Format Specification (DAFS) C-language
API. Sponsored by ARPA, DAFS has been designed to be the universal OCR
and document understanding format. The main feature of this release is
a greatly expanded Programmer's Guide and Reference.

The key points of DAFS:

International Capable (uses the Unicode standard)
Allows complex document structure
allows ambiguity and uncertainty
Compact, extensible, single file storage of text and image

The C-language API to DAFS, DAFSLib, allows one to manipulate DAFS
structures programmatically. The companion viewer/editor to DAFS is
called Illuminator and is an X-windows/Motif application. This latest
version of DAFS may be downloaded using this URL:


In this directory are the DAFS 0.9.2, and the complementary
0.9 release of Illuminator.


DAFS is a standard for OCR, not an OCR program. Many users have
requested that we supply an example OCR program that uses DAFS, but we
haven't had the resources to this point. If you have source to a free
OCR program, we would be interested in converting it to use DAFS.

There are many issues in running a Motif program like illuminator
under OpenWindows and Solaris. Don't give up! The TROUBLE_SHOOTING
file explains how to patch around them all.


( - old features
+ new features )

- sparc binaries
- linux binaries
+ sgi binaries

+ dafs2txt -- print out text
+ pda2dafs -- convert Calera PDA files
- dumpdafs -- dump out DAFS data in human readable form
- prall -- hierarchy text output
- sample files
+ bunch -- combine multiple DAFS files

+ updated manual


DAFS Lib is as free as we could make it. There are no royalties, no
restrictions on use, and only the disclaimer that RAF Technology and
ARPA provide this software with no warranty, and that you must give
credit to us. We are planning to release the source within the next
2-6 months with version 1.0. If there are any features missing in
DAFSLib that you would like to see, please let us know.

If you would like to use DAFS Lib or Illuminator but don't have a
Sparc, IRIX Workstation, or Linux PC, please let us know, and we may
be able to send you binaries for your machine.


The prime contractor is RAF Technology. The contact is:
This email address is being protected from spambots. You need JavaScript enabled to view it. . This software was developed for ARPA, under contract
MDA 904-93-C-7219.

Unicode is the emerging standard for International Character Encoding.
Unicode is Copyright The Unicode Consortium.

PDA is the document format of Calera Recognition Systems (now a part
of Caere), and is copyright by them.

TIFF was invented by Aldus.

A reminder: here is the CFP of the next DAS workshop:
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Jonathan Hull)
>Subject: CFP: 1996 IAPR Workshop on Document Analysis Systems

International Association for Pattern Recognition Workshop on
Document Analysis Systems
Malvern, Pennsylvania, October 14-16, 1996

Call for Papers

An intensive, three-day, single-track, 100% participation
workshop on the research and development of systems for the
analysis of document images.

Academic and industrial researchers, end-users and students are
encouraged to attend. We expect to provide several student scholarships

Attendance will be limited to 75 people. Papers will be delivered in a
single-track format. A portion of the workshop will also be devoted
to panel discussions and working groups. Participants will be invited
to propose topics for these in advance. There will be ample
opportunity for informal discussion and socializing.


IAPR Technical Committee on Graphics Recognition (TC-10)
IAPR Technical Committee on Text Processing (TC-11)
Loral Defense Systems-Eagan
Ricoh Corporation
U.S. Department of Defense

- descriptions of complete, working document image analysis systems,
including applications to large document image databases
- applications to text, graphics, maps, logic diagrams, music, etc.
- studies of system architectures
- methods of performance evaluation for whole systems or system modules
- interfaces to knowledge databases and post-processors
- algorithms for symbol recognition
- layout segmentation

Submission of Papers

Authors are invited to submit either a full paper or an abstract of
remarks. Full papers, of up to book-chapter length, will be presented
formally, published in a limited number of refereed Proceedings
available at the meeting, and considered for archival publication in a
post-workshop book. Abstracts of remarks, one page in length, will be
delivered more briefly, and will not be published in the Proceedings.
Authors should submit four copies of each paper or
abstract, cleanly typed and in English. If there are several authors,
please indicate the author that will handle all correspondence.
The email address of the corresponding author should be provided.

If any part of a full paper has been submitted to another conference,
please state where it has been submitted and how it differs
>from this paper.

Acceptance of a paper or abstract will result in automatic
distribution of registration materials.

Send full papers, abstracts, or inquiries about
the technical program to:

Jonathan J. Hull
Ricoh California Research Center
2882 Sand Hill Road, Suite 115
Menlo Park, CA 94025 USA
TEL: 415-496-5723 FAX: 415-854-8740 EMAIL: This email address is being protected from spambots. You need JavaScript enabled to view it.



Suzanne Liebowitz Taylor Loral Defense Systems-Eagan (USA)
Jonathan J. Hull Ricoh California Research Center (USA)

Program Committee

Henry Baird, AT&T Bell Laboratories (USA)
Horst Bunke, University of Berne (Switzerland)
Dick Casey, IBM Almaden (USA)
Andreas Dengel, German Research Center for Artificial Intelligence (Germany)
Stephen Dennis, U.S. Department of Defense
Hiromichi Fujisawa, Hitachi (Japan)
Tin Kam Ho, AT&T Bell Laboratories (USA)
Junichi Kanai, ISRI-UNLV (USA)
Louisa Lam, Concordia University (Canada)
Daniel Lopresti, Panasonic Technologies (USA)
Guy Lorette, IRISA, University of Rennes 1 (France)
Juergen Schuermann, Daimler Benz (Germany)
Larry Spitz, Daimler Benz (USA)
Sargur N. Srihari, CEDAR-SUNY/Buffalo (USA)
Karl Tombre, INRIA Lorraine & CRIN/CNRS (France)
Kazuhiko Yamamoto, Gifu University (Japan)

Important Dates

April 1, 1996 Submission of full Papers
June 15, 1996 Acceptance of full papers
July 1, 1996 Submission of abstracts
August 1, 1996 Acceptance of abstracts
August 15, 1996 Camera-ready copy of full papers
September 1, 1996 Registration payment and hotel reservation

Workshop Information

Pre-registration will be required.
Information about the workshop and the registration procedure
will be available over the World Wide Web at:


Other inquiries about local arrangements, registration,
and scholarships should be directed to:

Suzanne Liebowitz Taylor
Loral Defense Systems-Eagan
70 E Swedesford Road
Paoli, PA 19301-0517

TEL: 610-648-2009 FAX: 610-648-2288 EMAIL: This email address is being protected from spambots. You need JavaScript enabled to view it.

Here are some miscellaneous CFPs which may be of interest to some of
you. As some of you find my mails too long, I have started to shorten
some of the CFPs. But as an enhanced service, I give you access to my
database of CFPs, which is regularly updated at the following URL
>From: "Henrik I. Christensen" <This email address is being protected from spambots. You need JavaScript enabled to view it. >
>Subject: Call for Papers for Performance Characterisation Workshop

Announcement and Call for papers for:
Performance Characteristics of Vision Algorithms

Organizer: Patrick Stelmaszyk, ITMI, Grenoble
Programme Chairs: Wolfgang Foerstner, Bonn and
Henrik I. Christensen, Aalborg
Place: Cambridge, UK
Time: 19-20 April 1996.
Friday/Saturday (morning) after the ECCV '96

Program Committee:

Kevin Bowyer, USF Hans Burckhard, Hamburg Adrian Clark, Essex.
Patrick Courtney, ITMI Al Hanson, UMASS Bob Haralick, UW
Josef Kittler, UOS Chris McGlone, Pittsburgh Peter Meer, UoR
Keith Price, USC George Vosselman, Delft

The workshop is motivated by the strong belief that the lack of performance
characterization of vision algorithms is responsible for the hesitation of
industry to use Computer Vision as one of its tools. Reasons for this
situation are manifold: the lack of commonly accepted criteria for
evaluation, the lack of a methodology for testing, the lack of translating
the experience in testing of other engineering areas to Computer Vision and
possibly also the non-acceptance of empirical or theoretical comparisons of
vision algorithms, including their replication, as original research.

The goal of the workshop is to investigate the state of the art in
characterizing the performance of vision algorithms in all stages of the
development and use, such as the design of algorithms with a pre-specified
performance, the testing of algorithms with respect to give specifications
and the self-diagnosis of vision algorithms during their use in an
automated process.

We encourage authors to submit papers on the following topics:

* Theory and strategies for performance analysis of vision algorithms
* Linking analysis of vision experiments to the theory underlying the
* Characterization of the limitations of vision algorithms and/or
the class of image data for which a vision algorithm is suited or
not suited
* Demonstrations of the usefulness of performance characterization
and/or the limitations of statistical testing in Computer Vision
* Modularization of vision tasks and the characterization of networks of
vision algorithms

We will not consider plain comparisons of algorithms without explicit
relation to the underlying theory, plain statistical papers without
explicit relation to performance characterization of papers on the analysis
of the complexity of algorithms without containing a clear cost benefit

We intend to select approx. 4 papers for long (45 min.) and approx. 8
papers for short (30 min.) oral presentations forming the basis for a
common discussion. An additional 12 - 16 papers will be selected for
poster presentation possibly including software demos.


15. September 1995 Call for papers
1. December 1995 Full papers
15. January 1996 Reviews
1. February 1996 Notification of authors
15. March 1996 Camera ready manuscripts are due
1. April 1996 Distribution of papers to participants (ftp)
19. April 1996 Workshop (1.5 days)
31. May 1996 Distribution of Proceedings to participants

Submit Papers before 1. December 1995 to:

Dr. Henrik I. Christensen
Laboratory of Image Analysis
Aalborg University
Fr. Bajers Vej 7, Bldg. D1
DK-9220 Aalborg OE, Denmark
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Charles Dyer)
>Subject: CVPR '96


IEEE Computer Society Conference on
Computer Vision and Pattern Recognition

June 18 - 20, 1996
San Francisco, California, USA

1. Important Dates

OCTOBER 25, 1995 - Papers due
(NOTE: This is a new date, and is an absolute
deadline for receipt of papers)
February 5, 1996 - Notification of acceptance
March 25 - Camera-ready manuscripts due
June 18-20 - Conference
June 16-17, 21-22 - Workshops

2. Web Page
For complete and up-to-date information on CVPR '96, see the World-
Wide Web page at http://www.cs.wisc.edu/cvpr96/

3. The Program
The program will consist of high quality contributed papers on all
aspects of computer vision and pattern recognition. Original
(previously unpublished) manuscripts on topics in the following
categories are solicited:

Segmentation, Grouping
Physics-based Vision
Low-Level Processing
Texture Analysis
Pattern Analysis
Motion Analysis
Integration of Modules and Cues
Shape and Object Representation
Object Recognition
Learning in Computer Vision
Active and Real-Time Vision
Multi-Sensor Image Analysis

The conference is not limited to these topics. If a paper is not
exactly on a topic in one of these categories, it may be specified
that the paper falls in two or more categories, or that the paper is
perhaps best not categorized at all. The categories are designed to
closely match the papers with the expertise of the reviewers who will
be members of the Program Committee. Every category will be
represented by several Program Committee members. As is customary
with CVPR, reviewing will be double-blind.

4. Paper Submission Four (4) copies of manuscripts should be
received no later than OCTOBER 25, 1995 (this is a hard deadline) at
the following address:

Professor Katsushi Ikeuchi
Wean Hall 4212
Computer Science Department
Carnegie Mellon University
5000 Forbes Ave.
Pittsburgh, PA 15213-3891 USA

Telephone: (412) 268-6349

The form of the manuscript should be as follows (in this order):

(1) First Title Page
Contains title, author(s), postal address(es), e-mail
address(es), abstract up to 200 words, and a list of one or
more categories that best describe the topic of the manuscript.

(2) Second Title Page
Contains title and abstract only.

(3) Summary Page
Contains responses to each of the following questions (each
answered separately):
I. What is the original contribution of this work?
II. Why should this contribution be considered important?
III. What is the most closely related work by others and how
does this work differ?
IV. How can other researchers make use of the results of this
V. If any part of this work has been submitted to other
conferences, workshops or journals, please state where it has
been submitted and how it differs from this paper.

(4) The Manuscript
No more than 30 pages (double-spaced, 12-point font) including
all text, figures and references (but no title or abstract).

Double submission of manuscripts to CVPR and ECCV is permitted, but
double publication is not.


6. Email Mailing List
While all information relevant to the conference will be updated on
this home page, you can also subscribe to an email mailing list in
order to receive occasional updates from the Organizing Committee.
The name of this mailing list is CVPR96, which is managed by the
Listproc software package. Only the Organizing Committee is
authorized to send mail to this list. If you have any questions about
this list, send email to This email address is being protected from spambots. You need JavaScript enabled to view it. or to CVPR96-
This email address is being protected from spambots. You need JavaScript enabled to view it. .

To subscribe to this list, send email to This email address is being protected from spambots. You need JavaScript enabled to view it.
with the BODY of the message containing the line:

subscribe CVPR96 <your real name>

7. Affiliated Workshops

7.1. Function, Formation, and Facilitation Workshop June 17.
Organizers: Azriel Rosenfeld (This email address is being protected from spambots. You need JavaScript enabled to view it. ) and Ehud Rivlin
(This email address is being protected from spambots. You need JavaScript enabled to view it. )

7.2. Third CAD-Based Vision Workshop June 21 - 22. Organizers:
George Stockman (This email address is being protected from spambots. You need JavaScript enabled to view it. ), Octavia Camps
(This email address is being protected from spambots. You need JavaScript enabled to view it. ), and Pat Flynn (This email address is being protected from spambots. You need JavaScript enabled to view it. )

7.3. Mathematical Methods in Biomedical Image Analysis Workshop
June 21 - 22. Organizer: Amir Amini(This email address is being protected from spambots. You need JavaScript enabled to view it. )

>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Jeffrey J. Rodriguez)



April 8-9, 1996, Historic Menger Hotel, San Antonio, Texas, U.S.A.

You are cordially invited to attend the IEEE Southwest Symposium on
Image Analysis and Interpretation. The Symposium is sponsored by
The University of Texas at San Antonio, The University of Arizona, and
the Central Texas Section of the IEEE, in cooperation with the IEEE
Signal Processing Society.

The Southwest Symposium will again bring together researchers in
academia, industry, and government working in the area of image analysis.
The program will consist of high quality papers on all aspects of
image analysis and interpretation.

Topics of interest include, but are not limited to, the following:

* Features & invariants * Stereo image analysis
* Boundary detection * Biomedial image analysis
* Segmentation * Remote sensing
* Color image analysis * Automated inspection
* Multiresolution techniques * Document image analysis
* Multidimensional/multi-modality analysis * Systems & applications

University of Texas at San Antonio
This email address is being protected from spambots. You need JavaScript enabled to view it.

Oklahoma State University
This email address is being protected from spambots. You need JavaScript enabled to view it.

Jeffrey J. Rodriguez
The University of Arizona
This email address is being protected from spambots. You need JavaScript enabled to view it.

University of Texas at San Antonio
This email address is being protected from spambots. You need JavaScript enabled to view it.

Southwest Research Institute
This email address is being protected from spambots. You need JavaScript enabled to view it.


J. K. Aggarwal University of Texas at Austin
Nasir Ahmed University of New Mexico
Steve Beck Tracor Applied Sciences (Austin, TX)
Alan C. Bovik University of Texas at Austin
Dapang Chen National Instruments (Austin, TX)
Melba Crawford University of Texas at Austin
Larry Hanafy Texas Instruments (Dallas, TX)
Phil Hill SVS, Inc. (Albuquerque, NM)
W. K. Jenkins University of Illinois, Urbana-Champaign
Nasser Kehtarnavaz Texas A&M University
Alireza Khotanzad Southern Methodist University
John C. Krumm Sandia National Laboratories (Albuquerque, NM)
Murray H. Loew George Washington University
Sunandra Mitra Texas Tech University
David C. Munson University of Illinois, Urbana-Champaign
William O'Brien University of Illinois, Urbana-Champaign
K. R. Rao University of Texas at Arlington
Scott Starks University of Texas at El Paso
Robin N. Strickland The University of Arizona
Monte Tull AT&T (Oklahoma City, OK)
David Waits Oklahoma State University
Rabab Ward University of British Columbia (Canada)


To propose a paper, prepare a summary of the paper using a 12-point font
size and no more than three pages, including figures and references.
Send four copies of the paper summary, along with a cover sheet stating the
paper title, technical area(s), and corresponding author's information
(telephone, FAX, mailing address, e-mail address) to

Prof. Scott T. Acton
School of Electrical & Computer Engineering
202 Engineering South
Oklahoma State University
Stillwater, OK 74078 U.S.A.

Each accepted paper will be published in the conference proceedings,
conditional upon author's advance registration. A LaTeX style file for
preparing the final paper will be provided for the authors' convenience.


November 2, 1995: Submissions of three-page summary
January 5, 1996: Notification of acceptance
February 2, 1996: Final camera-ready paper & registration due

FURTHER INFORMATION: http://www.ece.arizona.edu/conferences/swsymp96

>From: C Bouman <This email address is being protected from spambots. You need JavaScript enabled to view it. >
>Date: Wed, 6 Sep 1995 22:51:26 -0500

Dear Image and Multidimensional Signal Processor,

The Ninth IMDSP Workshop will be held in Belize on March 3-6, 1996. In
addition to providing an intimate setting for interaction with your IMDSP
colleagues, Belize offers unparalleled opportunities for recreation,
ranging from snorkling on coral reefs to exploration of spectacular rain
forests and Mayan ruins. Belize is a British Commonwealth country with a
stable, democratic government. Most Belizeans speak English.

I am writing to encourage you to submit a paper to this workshop. In
response to complaints that our original submission date came at a very
difficult time, we have extended the deadline to October 1, 1995. In
addition, for those who are concerned about the difficulty and cost of
reaching Belize, we have the following information:

Economy airfare from continental U.S. (Saturday stay required)
Lodging (Sunday - Tuesday, including all meals, based on single occupancy)
Approximate registration

Total cost

Belize City is approximately a 2 hour flight from Los Angeles, Houston, or

The Call for Papers follows. More information about the workshop and
Belize can be found at our web site. The URL is:


Don't miss this unique event. Plan to join us in Belize in March.


Jan Allebach and Jelena Kovacevic, General Co-chairs


IEEE Signal Processing Society
IS&T The Society for Imaging Science and Technology
Ninth IMDSP Workshop
March 3-6, 1996. Radisson Hotel
Belize City, Belize

General Co-Chairs:

Jan P. Allebach
Purdue University

Jelena Kovacevic
AT&T Bell Laboratories

Program Chair:

Edward J. Delp
Purdue University

The Image and Multidimensional Signal Processing Technical Committee of the
IEEE Signal Processing Society has established a tradition of biennial
workshops that have been intellectually stimulating and held in interesting
places. The Ninth Workshop, to be held in cooperation with the Society for
Imaging Sceince and Technology continues this tradition. It will be held in
Belize which is located on the Caribbean Sea directly south of the Yucatan
Peninsula and east of Guatemala, just two hours by plane from Miami. Belize
has a subtropical climate. The workshop will take place in the dry season
when temperatures will average 22C. There are beautiful Mayan ruins in
Belize and the longest barrier reef in the Western Hemisphere. The workshop
will be held in Belize City at the Radisson Hotel.

The theme of the workshop is ``Model-Based Methods in Imaging: From Content
to User.'' This was partly motivated by the NSF-sponsored workshop on
Future Directions in Signal Processing that was held in Keystone, Colorado
in November 1991. At this workshop, the task group in the IMDSP area
identified a need for more emphasis on model-based methods, including
models for signal content and models for the human viewer. In addition, it
was felt that there was a need for more work on image quality metrics. It
is the intention of this workshop to address these topics:

Physics-based models
Multiresolution models
Models for the human visual system
Color models
Image and video quality metrics
Image and video compression
Image and video analysis
Image rendering for printing and display
Color imaging
Radar imaging
Medical imaging

Participation in the workshop will be both by invitation and application.
Prospective attendees will submit a 2-page summary in final form. If the
paper is accepted the summary will appear in the workshop digest as
submitted. Authors should obtain the author's kit via anonymous ftp to
skynet.ecn.purdue.edu ( in the directory /pub/dist/imdsp96.
The author's kit provides complete information relative to the submission
process. It is the intention of the Committee to conduct all communication
with authors electronically. Questions should be directed to
This email address is being protected from spambots. You need JavaScript enabled to view it. . The General Co-Chairs can be reached at:

Jan Allebach
+1 317 494 3535

Jelena Kovacevic
+1 908 582 6504

Authors' Schedule
October 1, 1995 Submission of two page summaries
October 15, 1995 Notification of acceptance
November 15, 1995 Registration due with deposit

[Rest of CFP deleted -- see full text in my archives]
>From: yang+@cs.cmu.edu (Jie Yang)
>Subject: Call for Paper


First International Conference on Multimodal Interface

Beijing, China
March 29-31, 1996

Sponsored by
Chinese Association of Automation(CAA)
Chinese Computer Federation(CCF)
Center for Hi-Tech Development(CHiTD), SSTCC
National R&D Center for Intelligent Computing System(NCIC)

In Cooperation with
IEEE Computer Society

A new generation of interactive multimodal systems is beginning to
emerge with advanced technology of spoken language understanding,
image processing, character recognition, natural language processing.
By this technology, new interfaces are designed to permit people to
speak and write in their native language, to gesture, draw, and
point while speaking, to control device via human gaze, to interact
with 2D or 3D visual displays involving complex graphics, maps,
videos, and animated human faces, and receive feedback via speech
and non-speech audio, tactile, and other media. ICMI aims at to
unite people from R&D and industry for an exchange on a wide variety
of topics related to this new area.

Submissions are invited on original and substantial academic and/or
industrial research on all aspects of multimodal interface.
Interested topics include:
A. perceptual computing model and system;
B. speech recognition;
C. text to speech & speech to text;
D. character recognition;
E. handwriting and printed document to text;
F. face, hand gesture, lip-motion recognition;
G. hand gesture to text & text to hand gesture;
H. natural languages and lip-motion, facial expression, hand gesture;
I. machine translation and multi-language interface;
J. multimodal interface and computer graphics;
K. image retrieval, data compression and multimedia interface;
L. platform & tools;
M. multimodal interface in internet;
N. multimodal interface in virtual reality
O. other applications;

Dates to Note:
o Nov 15, 1995 -- Paper submission deadline
o Dec 15, 1995 -- Notices mailed to accepted/rejected authors
o Jan 31, 1996 -- Final (camera ready) papers due
o Mar 29 - 31, 1996 -- Conference dates

Prospective authors are invited to submit papers relevant to
multimodal interface in any of the conference interested topics.
Papers must be received by the ICMI'96 Program Committee no later
than November 15, 1995. Papers will be selected by the ICMI'96
Program Committee and assigned for oral presentation. English is
the working language for the conference. Submission of a paper
implies a commitment to submit an eight (or less) page,
camera-ready version of the paper and to present the paper in
technical session if the paper is accepted.

Accepted submissions will be published in a proceedings volume.
Extended versions of selected high-quality papers will be
submitted to the Journal of Computer Science and Technology after
the conference.


Submissions should be no longer than 8 pages single spaced
(excluding title page and references), and should use a font no
smaller than 12-point. The title page should include the title of
the paper, a short abstract (200 words), a list of keywords,
at most three choices of technical area(s) indicated in order of
preference using the codes (A - O), and the authors' postal and
email addresses and fax and phone numbers.

Mail 4 copies of the submission paper to:
Dr. Wen Gao
ICMI'96 program committee
National R&D Center for Intelligent Computing System,
P.O. BOX 2704, Beijing 100080
email:This email address is being protected from spambots. You need JavaScript enabled to view it.

For further information about the submission, contact:
Ms. Mei Du,
National R&D Center for Intelligent Computing System,
P.O. BOX 2704, Beijing 100080
email:This email address is being protected from spambots. You need JavaScript enabled to view it.


An event for industry and research groups to show the new products and
new prototypes and application systems is scheduled during the
conference in the conference site. For further information about the
ICMI'96 exhibition, contact:
Minghong Shen, Exhibition Chair of ICMI'96
3rd floor,
No.7, NanHeng Street,
SanLiHe Rd.
BeiJing, 100862
Tel: +86-10-8513677 Ext.3303 or 3305
Fax: +86-10-8511858

[Rest of CFP deleted -- full copy in my archives]
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Manuel Trucco)
>Subject: CFP: Seventh British Machine Vision Conference



9-12 September 1996

The University of Edinburgh

Robert B Fisher Emanuele Trucco
University of Edinburgh Heriot-Watt University
This email address is being protected from spambots. You need JavaScript enabled to view it. This email address is being protected from spambots. You need JavaScript enabled to view it.

Gillian M Hayes Paul Siebert
University of Edinburgh The Turing Institute
This email address is being protected from spambots. You need JavaScript enabled to view it. This email address is being protected from spambots. You need JavaScript enabled to view it.

The British Machine Vision Conference is the main UK conference for machine
vision and related topics. High quality contributed papers are sought
describing recent and novel research in the areas of computer vision,
image analysis and processing or pattern recognition. Papers describing
research being undertaken throughout national or international collaborative
projects are particularly welcome. The conference is a single-track meeting
with both oral and poster presentations. In addition to the contributed papers,
there will be talks by invited speakers and a pre-conference tutorial
programme (free to registered students).


Contributions are sought on any novel aspect relating to machine vision and
pattern analysis, including:

* image processing and feature extraction
* object recognition and scene analysis
* reconstruction of 3D shape
* computational issues in perception
* robotic vision and sensory fusion
* advanced pattern analysis
* practical applications
* model-based coding
* architectures
* active vision
* motion analysis
* neural network

Four copies of full papers not exceeding 10 pages (approx. 5000 words if no
figures), three of which anonymous, should be submitted for review.
Please include two title pages, each containing title, authors' names
(on one copy only) and an abstract. Papers will be reviewed by the elected
BMVA Committee with a small number of specially coopted members.
By submitting a paper to BMVC, the author(s) warrant that it, and any related
paper with essentially the same technical content, has not been and will not
be submitted to any other conference during the BMVC review period.
Papers will be accepted either for oral presentation or for presentation as
posters. A printed copy of the Proceedings will be available to delegates at
the conference and a selection of the best papers will be published separately
in a special issue of 'Image and Vision Computing'.

Guidelines for the format of papers can be found on the PEIPA at:
FTP: peipa.essex.ac.uk in ipa/info/conferences/bmvc (login: anonymous)
WWW: http://peipa.essex.ac.uk (under BMVC96)

Papers should be submitted to the Conference Secretary:
Mrs Judith Gordon
Department of Artificial Intelligence
The University of Edinburgh
5 Forrest Hill
Edinburgh EH1 2QL
Tel: +44 131 650.3094
Fax: +44 131 650.6899
Email: This email address is being protected from spambots. You need JavaScript enabled to view it.

| |
| Deadline for paper submission 29 April 1996 |
| Notification of acceptance 10 June 1996 |
| Deadline for camera-ready copy 8 July 1996 |

The British Machine Vision Association and Society for Pattern Recognition
is a Company limited by guarantee, No.2543446, registered in England and Wales.
Registered Office: Granta Lodge, 71 Graham Road, Malvern, Worcs. WR14 2JS.
The Association is a non-profit-making body and is registered as charity
No. 1002307.

>From: Rachid Deriche <This email address is being protected from spambots. You need JavaScript enabled to view it. >




26-28 juin 1996
June 26-28, 1996
Paris, France

Organisee par le CEREMADE et l'INRIA/Organized by CEREMADE and INRIA


This conference is devoted to image processing, wavelets and partial
differential equations. Its aim is to discuss the impact on image analysis
of the recent mathematical developments of mathematical theories of
multiscale analysis (wavelet theory and variants : wavelet packets,...),
partial differential equations, variational methods that have recently
been proposed for image processing tasks like image restoration, shape
recognition, etc... Presentations by image processing researchers and
by mathematicians on both theoretical and practical aspects will be
This conference is the 12th in the series of International Conferences
on Analysis and Optimization of Systems organized by INRIA and is the
third specialized conference.



Methods Image applications
------------ ---------------------

Variational methods Image restoration
Multiscale analysis Segmentation
Geometric PDE's Shape modelling
Wavelets Motion analysis
Adaptive basis Computer vision
etc... Medical and satellite imagery


Authors are invited to submit original and unpublished contributions.
These contributions may focus on theoretical results as well as on
illustrative applications. The full contribution (7 pages) with key words,
abstract and e-mail address should be submitted in 5 copies to the conference
secretary. The proceedings will be published by Springer-Verlag.


15 novembre/November 15, 1995
- ------------------------------
Date limite de soumission des communications completes/deadline for
submission of full papers

31 janvier/January 31, 1996
- ----------------------------
Notification de l'acceptation aux auteurs/ notification of acceptance

20 mars/March 20, 1996
- ------------------------
Communication complete prete a etre imprimee, et la version electronique du
texte/ camera ready paper and electronic version of papers


Secretariat de la conference/Conference Secretariat

INRIA Rocquencourt
Relations Exterieures
Bureau des Cours et Colloques
BP 105
78153 Le Chesnay Cedex

Tel : 33 1 39 63 56 75
Fax : 33 1 39 63 56 38
email : This email address is being protected from spambots. You need JavaScript enabled to view it.

>From: Andre Gagalowicz <This email address is being protected from spambots. You need JavaScript enabled to view it. >
>Subject: IMAGECOM'96
3rd International Conference

Co-located with

First European Conference on


Palais des Congres Palatium d'Arcachon
20 - 24 May, 1996

Call for papers

Supported by :

European Union : ACTS, COST 237

EBU : European Broadcasting Union

Organised by

SEE : Societe des Electriciens et des Electroniciens
IREST : Institut de Recherches Economiques et Sociales sur les
ADERA : Association pour le Developpement de l'Enseignement et des
Recherches aupres des Universites, des Centres de Recherche et des
Entreprises d'Aquitaine

With the backing of
France Telecom


The last few years have seen technological and normative breakthroughs
in the field of image computing and graphics as well as in multimedia
networking. New perception and recognition techniques close to
robotics are being developed. These, together with convergence of
image creation, analysis and representation techniques, are leading to
new developments in multimedia production and services. The access to
digitised image, sound and data from the workplace and from the home
relies on increasingly smarter bearer services. As added value is
transferred towards production and services, strategic alliances are
being built to take advantage of the multiplication of the
distribution channels and media. User access is being made easier,
more intuitive, by means of smarter, knowledge-based approaches. Fast
retrieval of essential information and browsing ever increasing
amounts of information require smart intelligent guides, software and
agents, as well as multimodal interfaces. Correspondingly, adequate
development of network infrastructure, architecture and protocols is

This third issue of the IMAGE'COM is co-located with the first ECMAST
(European Conference on Multimedia Applications, Systems and
Technologies) to provide a forum to monitor and anticipate the
developments that make multimedia applications possible, to exchange
information and results on recent work, and to help
cross-fertilisation of image and communication research activities.
The following is a non exhaustive list of topics addressed.

1. Production, creation
Acquisition, pre-processing, knowledge extraction and recognition
Computer perception, signal sensing, active vision
Equipment and software for image analysis and synthesis

2. Advanced coded representation of image, sound and data
Knowledge-based processing, content-based representation
Advanced audio-visual functionalities
Compression, coding, very low bit rate coding
Computer graphics
Virtual reality

3. Networking
Core network architectures
Access network architectures (local loop)
Return channels
Service multiplexing

4. Servers, terminals and storage
Mobility and multimedia
Multimedia data bases
Terminal equipment architecture
Information servers
Hypermedia browsing
5. Services
Contents for multimedia services
Pilot applications, agent applications, intermediation
Conditional access, billing, customer care, service management
Service users and ergonomics, usability of terminals,
man-machine interfaces.


Those who wish to contribute to one of these themes should send their
typed summary in English (French translation optional) of between 600
to 1000 words and one or two illustrations, to the conference
secretariat by September 30, 1995 It would be appreciated if the
authors could mention 3 or 4 keywords.


Authors must produce a text in English accompanied by a summary in
French if desired. Authors will be informed by the end of December
1995 of the Programme Committee's selection from papers submitted.
For the selected contributions, a full length camera ready copy in
English will be required by March 30,1996. The selected papers will
be published in the conference proceedings at the organisers'

October 31, 1995: (was previously September 30,1995)
deadline for summaries.
December 30, 1995:
acceptance of papers
March 30, 1996:
deadline for full papers


Authors should provide
their complet address
(including fax)
and send their summary to :
BP 196
Phone (33) 56 15 11 57
Fax (33) 56 15 11 60

Now to some Advance Programs and registration information for coming
events. Soe of you have asked me to do some more editing to shorten
the TC10 mailouts. So Most of the time I only keep essential
information. Here also, i have an archive of the full things, which
you can access at URL http://www.loria.fr/~tombre/Programs/maillist.html
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (RB. Johnson)
>Subject: IEE/BMVA DIP Colloquium - 2 November 1995



Colloquium organised by IEE Professional Group E4 (Image
processing and vision) and co-sponsored by the British
Machine Vision Association to be held at Savoy Place,
London, UK on

Thursday 2 November 1995
[Full program in the archives]

For further information and application forms please contact:

Ms Janet McCready (Groups Officer)
The Institution of Electrical Engineers
Savoy Place
London WC2R 0BL
United Kingdom

Tel: +44 (0) 171 344 5421
Fax: +44 (0) 171 497 3633
E-Mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Lee_Tong)
>Subject: ICSC'95 Call for Participation


ICSC '95 - Third International Computer Science Conference:
Image Analysis Applications and Computer Graphics
Nikko Hotel, HONG KONG, 11-13 December, 1995

Organised and Sponsored by the IEEE Hong Kong Section, Computer Chapter.

In cooperation with IEEE Computer Society, International Association for
Pattern Recognition (IAPR), ACM (HK).

ICSC '95 is the Third International Conference on Computer Science to be
held in Hong Kong. Its purpose is to provide a forum for scientific
interchange among computer scientists and engineers from all over
the world, especially from the Asia-Pacific region. This conference
will focus on a broad spectrum of research topics related to
image analysis applications and computer graphics.

The proceedings will be published in the Springer-Verlag Lecture Notes
in Computer Science series in time for distribution at the conference.


DAY ONE: Prof. T. S. Huang
Beckman Institute for Advance Science and Technology
University of Illinois at Urbana-Champaign
U. S. A.

Title: 3D MODEL-BASED VIDEO CODING: Computer Vision Meets Computer Graphics

[abstract and biography deleted]


DAY 2: John Lasseter
Director & Animator
Vice President, Creative Development, Pixar, U.S.A

Title: Creating Memorable Characters with Computers

[biography deleted]


Image Analysis Applications Computer Graphics

Robot Navigation and Tracking Scientific Visualization
Feature Extraction Geometric Modelling
Document Processing Curves & Surfaces

3D Image Analysis Rendering
Biomedical Imaging Image Synthesis

Application Systems Simulation and Animation
Optical Character Recognition Human Models


Title: Document Image Analysis Techniques: An Overview

Prof. Rangachar Kasturi, Professor of computer science and engineering,
Pennsylvania State University, U.S.A.

[abstract and biography deleted]


Day 1 Day 2 Day 3

8:45- 9:00 Opening Remarks

9:00-10:40 IA1a / CG1a IA2a / CG2a IA3a / CG3a

10:40-11:00 Break Break Break

11:00-12:00 Invited talk Invited talk IA3b / CG3b (11:00-12:20)
T S Huang J Lasseter

12:00- 1:50 Lunch Lunch Lunch (12:20-1:30)

1:50- 3:30 IA1b / CG1b IA2b / CG2b Tutorial (1:30-5:00)
3:30- 3:50 Break Break |
3:50- 5:10 IA1c / CG1c IA Posters V

Reception Banquet


[Full program deleted]

[Registration deleted]
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Rohini Srihari)
>Subject: AAAI Fall Symp


AAAI 1995 Fall Symposium Series
Computational Models for Integrating Language and Vision

Nov 10-12, MIT, Cambridge, MA

Note: Registration closes on Oct 15, 1995.
For registration information, see
For more information on this symposium, see

Friday, Nov 10

9:00--10:30 Opening Session: Methods and Issues
Survey of Topic -- Rohini Srihari, CEDAR, SUNY at Buffalo
Survey, Part 2 -- To be announced

10:30--11:00 Break

11:00--12:30 Paper Session: Event Perception/Description
Session Chair: Jeff Siskind

12:30--2:00 Lunch

2:00--3:30 Paper Session: Gesture Understanding
Session Chair:

3:30--4:00 Break

4:00--5:30 Session: Knowledge Representation for Language, Vision
Session Chair: Candace Sidner

6:00--7:00 Reception

Saturday, Nov 11

9:00--10:30 Session: Spatial Reasoning
Session Chair: Annette Herskovits

10:30--11:00 Break

11:00--12:30 Invited Talk: Joe Mundy, GE Corporate Research Division
Image Observation Events: A Link to Linguistic Context

12:30--2:00 Lunch
2:00--3:30 Session: Image/Diagram Understanding
Session Chair: Gordon Novak

3:30--4:00 Break

4:00--5:30 Panel: Hard problems, where to focus research, impact of
Panel: To be announced

7:30--9:00 Public Forum (10 min overview of each symposium)
Speaker: Ken Forbus

Sunday, Nov 12

9:00--10:30 Session: Image/Video Indexing
Session Chair:

10:30--11:00 Break

11:00--12:30 Wrapup and Discussion
Summary of issues --
Final discussion (Moderator: Rohini Srihari)

[All this shortened -- see full version in my archives]

Well, despite all this shortening, this mailout was still a very long
one. I hope you found some useful information...

Your TC10 chairman,

Karl Tombre - INRIA Lorraine & CRIN-CNRS --- Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
Post: 615 rue du jardin botanique, BP 101, 54602 Villers CEDEX, France --or--
Batiment LORIA, BP 239, 54506 Vandoeuvre CEDEX, France
Phone: +33 --- Fax: +33 --or-- +33
WWW URL: http://www.loria.fr/~tombre