Newsletter 25


- Announcement: "Document Image Analysis" book published by IEEE
Computer Society Press
- Announcement: Job opening in algorithm and software development in
document image analysis
- Conference schedule: SDAIR'95
- CVL technical report archive
- CFP: ICSC'95
- Final announcement: statistical shape analysis
- First announcement: ECCV'96
- CFP: workshop on vision for robots

Dear IAPR TC10 friends. This TC10 news is just for sending out a
number of various announcements I have received or collected since the
last mailout. The next TC10 news should give you some fresh
information on the coming Graphics Recognition workshop, and maybe
also on ICDAR. Also, if some of you missed the previous information
posted about the dashed-line contest, feel free to ask me for a copy
of this information.

The book by L. O'Gorman and R. Kasturi about document image analysis
is now available. Here is the information about it. I would like to
recommend it; I ordered it and I liked it very much, it is a nice
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Larry O'Gorman)

"Document Image Analysis," by Lawrence O'Gorman of Bell Labs, Murray
Hill, and Rangachar Kasturi of Pennsylvania State University is a new
book covering methods and applications in the document image
analysis field. The book includes tutorial material describing problem
areas and appropriate methods. This is supplemented by the
inclusion of important papers in the field that detail the subjects.
Some described methodologies are: image binarizarion, noise reduction,
line and curve fitting, shape recognition, skew estimation, page
layout analysis, handwritten character recognition, and
machine-printed character recogntion. Some applications are: postal
address recognition, map recognition, electronic library analysis, tax
form reading, and music recognition.

The book is published by the IEEE Computer Society Press, 1994,
$44.00, ($33.00 IEEE member price), IEEE catalog number EH0410-1,
ISBN 0-8186-6547-5.


I also received this announcement about a job opening in document
image analysis from NYNEX.

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

Job Announcement

We need to fill a position in algorithm and software development
in document image analysis. Specifically, the job involves working
with a small team to develop and code algorithms for line drawing
interpretation. The position needs to be filled as soon as possible.

o MS in CS, EE, or related field
o Pertinent experience in document image analysis
o Excellent C and Unix skills
o Excellent communication skills

o PhD in CS, EE, or related field
o Experience in line drawing interpretation, specially interpretation
of utility company drawings
o Experience in hand-printed or machine-printed character recognition
o Experience working in a team of software developers using source
code control
o Experience with SunOS and Solaris
o Knowledge of X/Motif

NYNEX is the Regional Bell Operating Company or RBOC (the local phone
company) for most of New York and New England. NYNEX Science & Technology
is the R&D division of NYNEX.

Applicants should send their resumes by mail, fax, or e-mail to the
address below.

Atul K. Chhabra Phone: (914) 644-2786
Member of Technical Staff Fax: (914) 644-2404
NYNEX Science & Technology Internet: This email address is being protected from spambots. You need JavaScript enabled to view it.
500 Westchester Avenue
White Plains, NY 10604


The SDAIR'95 symposium had been announced in this mailing list. The
conference program has now been decided upon and here it comes:

>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Andrew D Bagdanov)

Fourth Annual Symposium
on Document Analysis
and Information Retrieval

April 24 - 26, 1995

Desert Inn Hotel
Las Vegas, Nevada

Sponsored by the

Information Science Research Institute
The Howard R. Hughes College of Engineering

University of Nevada, Las Vegas



Sunday, April 23, 1995

7:00pm - 10:00pm Desert Inn
Reception and Registration

Monday, April 24, 1995

7:00am - 8:20am Desert Inn

8:20am - 8:30am Desert Inn

Donna Harman, Chairperson
National Institute of Standards and Technology

William R. Wells, Dean
Howard R. Hughes College of Engineering
University of Nevada, Las Vegas

Thomas A. Nartker, Director
Information Science Research Institute
University of Nevada, Las Vegas

8:30am - 9:15am Desert Inn
Invited Speaker

Spoken Document Retrieval - A Multimedia Tool
Karen Sparck Jones
University of Cambridge, United Kingdom

9:15am - 10:30am Desert Inn
Session 1

Automatic Extraction of Information From Printed Documents
Gregory K. Myers, Prasanna G. Mulgaonkar;
SRI International

Dynamic Hypertext Links for Highly Degraded Data in TELLTALE
Claudia Pearce; U. S. Department of Defense

Space and Time Improvements for Indexing in Information Retrieval
Willie Rogers, Gerald Candela, Donna Harman;
National Institute of Standards and Technology

10:30am - 11:00am Desert Inn
Refreshment Break

11:00am - 12:40pm Desert Inn
Session 2

Textual Database Lexicon Used as a Filter to Resolve Semantic
Ambiguity Application on Multilingual Information Retrieval
Khaled Radwan, Christian Fluhr; Institut National des
Science et Techniques Nucleaires, France

Domain Terminology Patterns in Different Disciplines: Evidence
From Abstracts
Stephanie W. Haas; University of North Carolina at Chapel Hill

Corpus-Specific Stemming Using Word Form Co-Occurrence
W. Bruce Croft, Jinxi Xu; University of Massachusetts, Amherst

Information Retrieval Based on Word Senses
Hinrich Schutze, Jan O. Pedersen;
Xerox Palo Alto Research Center

12:40pm - 1:45pm
Lunch Desert Inn

1:45pm - 2:30pm Desert Inn
Invited Speaker

Character Segmentation in Document OCR: Progress and Hope
Dr. Richard G. Casey
IBM Almaden Research Center (Retired)

2:30pm - 3:45pm Desert Inn
Session 3

Improving OCR Performance With Word Image Equivalence
Tao Hong, *Jonathan J. Hull; State University of New York
at Buffalo, *RICOH California Research Center

A Word-Level Deciphering Algorithm for Degraded Document Recognition
Chi Fang, *Jonathan J. Hull; State University of New York
at Buffalo, *RICOH California Research Center

The Recognition of Noisy Polyfont Printed Text Using Combined HMMs
Andrew J. Elms, J. Illingworth; University of Surrey,
United Kingdom

3:45pm - 4:30pm Desert Inn
Poster Session

Combining Statistics and Heuristics For Language Identification
Erhan Gokcay, Didem Gokcay; University of Florida

Functional Decomposition of Business Letters
Mark Lipshutz, Suzanne Liebowitz Taylor; Unisys Corporation

Generating Synthetic Data for Text Analysis Systems
David Doermann, Shee Yao; University of Maryland

Key-Relation Technology for Text Retrieval
Tadanobu Miyauchi, Mamiko Oka, Yoshihiro Ueda; Fuji Xerox

Lexical Assistance at the Information-Retrieval User Interface
Roy Byrd, Yael Ravin, John Prager;
IBM T. J. Watson Research Center

Query and Retrieval in Multimedia Databases: A New Perspective
Augusto Celentano; Politecnico di Bari

Representing Dimensioning Annotation in Engineering Drawings Using
Flat Matrix Grammar
Dov Dori; Israel Institute of Technology

Retrieving Captioned Pictures Using Statistical Correlations and a
Theory of Caption-Picture Co-Reference
Neil C. Rowe; U.S. Naval Postgraduate School

Using Natural Language Processing for Identifying and Interpreting
Tables in Plain Text
Shona Douglas, Matthew Hurst, David Quinn;
University of Edinburgh, United Kingdom

4:30pm - 5:45pm Desert Inn
Session 4

Validation and Estimation of Document Degradation Models
Tapas Kanungo, *Henry S. Baird, Robert M. Haralick;
University of Washington, *AT&T Bell Laboratories

Perfect Document Layout Ground Truth Generation Using DVI Files
and Simultaneous Word Segmentation from Document Images
Su Chen, Robert M. Haralick, *Ihsin T. Phillips;
University of Washington, *Seattle University

Efficiently Identifying the Faces of a Wireframe Reconstructed From
Three-View Engineering Drawings
Mu Hsing Kuo, R. E. Marston;
University of Nottingham, United Kingdom

6:00pm - 10:00pm
Happy Hour Great Hall
Dinner Thomas Beam
Tour of Facilities Engineering Bldg.

Tuesday, April 25, 1995

7:30am - 8:20am Desert Inn

8:20am - 8:30am Desert Inn

Donna Harman, Chairperson
National Institute of Standards and Technology

8:30am - 9:15am Desert Inn
Invited Speaker

New Directions in Clustering
Dr. Abraham Bookstein
University of Chicago

9:15am - 10:30am Desert Inn
Session 5

Recognition of Tables Using Table Grammars
E. Green, M. Krishnamoorthy; Rensselaer Polytechnic Institute

Schema-Independent Retrieval from Heterogeneous Structured Text
Charles L. A. Clarke, Gordon V. Cormack, Forbes J. Burkowski;
University of Waterloo, Canada

A Retrieval System for Graphical Documents
Oliver Lorenz, Gladys Monagan; Swiss Federal Institute of
Technology, Switzerland

10:30am - 11:00am Desert Inn
Refreshment Break

11:00am - 12:40pm Desert Inn
Session 6

Text Categorization of Low Quality Images
David J. Ittner, David D. Lewis, David D. Ahn;
AT&T Bell Laboratories

A Neural Network Approach to Topic Spotting
*&**Erik Wiener, *Jan Pedersen, **Andreas S. Weigend;
*Xerox Palo Alto Research Center, **University of
Colorado at Boulder

Text Categorization of Scanned Documents Applying a Rule-Based Approach
Claudia Wenzel, Rainer Hoch;
German Research Center for Artificial Intelligence (DFKI)

Text to Information: Sampling Uncertainty in an Example From
Physician/Patient Encounters
Timothy R. Thomas, Charlotte Kruger, Clint Scovel,
Joseph Shumate; Los Alamos National Laboratory

12:40pm - 1:45pm
Lunch Desert Inn

1:45pm - 2:30pm Desert Inn
Invited Speaker

Full-Text Search and Document Recognition of Japanese Text
Dr. Hiromichi Fujisawa
Central Research Lab, Hitachi, Ltd.

2:30pm - 3:45pm Desert Inn
Session 7

Structuring a Large Lexicon for Word Recognition
Chien Huei-Chen; SRI International

Document Reconstruction: A Thousand Words From One Picture
Jeffrey C. Reynar, *A. Lawrence Spitz, *Penelope Sibun;
University of Pennsylvania, *Fuji Xerox Palo Alto Laboratory

Neural Network Classifiers for Optical Chinese Character Recognition
Richard Romero, Robert Berger, Robert Thibadeau,
David Touretzky; Carnegie Mellon University

3:45pm - 4:00pm Desert Inn
Refreshment Break

4:00pm - 4:50pm Desert Inn
Session 8

An Adaptive Metaclassifier for Word Recognition Based on Multiple
Independent Classifiers
Thorsten Jaeger, Frank Hones, Andreas Dengel;
German Research Center for Artificial Intelligence (DFKI)

Evaluation of OCR Accuracy Using Synthetic Data
Tin Kam Ho, Henry S. Baird; AT&T Bell Laboratories

Wednesday, April 26, 1995

8:20am - 8:30am Desert Inn
ISRI Welcome

Thomas A. Nartker, Director
Information Science Research Institute
Howard R. Hughes College of Engineering
University of Nevada, Las Vegas

8:30am - 9:45am Desert Inn
The Fourth Annual Test of OCR Accuracy
ISRI Staff

9:45am - 10:00am Desert Inn
Refreshment Break

10:00am - 12:00pm Desert Inn
ISRI Research Reviews
ISRI Staff


Invited Speakers

Karen Sparck Jones is Reader in Computers and Information at the Computer
Laboratory, University of Cambridge. She has worked in automatic information
and language processing since the late fifties, directing a range of research
projects, and is the author of numerous publications in these areas. Her
research in the last decade has been in document indexing and retrieval,
database query, user and agent modelling, summarizing, and, language system
building and evaluation. She has also been professionally involved in
Cambridge, national and international activities and programmes, especially
in the language processing area. She is currently President of the
Association for Computational Linguistics, is a Fellow of the American
Association for Artificial Intelligence, and has received two awards for her
information retrieval research.

Richard G. Casey received the Doctor of Engineering Science degree from
Columbia University in 1965. His dissertation was in the area of character
recognition. From 1963 to 1970, he worked for the IBM Research Division at
Yorktown Heights. At Yorktown Heights, he worked with many of the pioneers
in pattern recognition and contributed to many of IBM's early recognition
products. He was transferred to the Almaden Research Center in San Jose in
1970 where he helped organize IBM's document recognition research. He has
received three "Outstanding Contribution" awards for research in OCR and forms
extraction which has led directly to several IBM products. He was an Invited
Speaker at ICDAR'91 in St. Malo, France. Since 1992, he has served on the
Advisory Board at ISRI and was program chairman for SDAIR'93. He retired from
IBM in 1994 and has spent the last six months at Ecole Nationale Superieure
des Telecommunications in Paris, working on a new book in document analysis.

Abraham Bookstein is a Professor of Information Science in the Psychology
Department of the University of Chicago. His education includes degrees in
Physics from the City College of New York, University of California (Berkeley)
and Yeshiva University (New York). His research, appearing in over 100
publications, has involved the application of mathematical models to problems
in Information Control, including statistical properties of information
generation and use, research methods, automated information retrieval, and,
most recently, data compression and coding. He has been the recipient of
numerous grants, including NSF awards and a Fullbright Fellowship. He was
the recipient of the American Society for Information Science Research Award.

Hiromichi Fujisawa received his B.E., M.E., and Ph.D. degrees in Electrical
Engineering from Waseda University in 1969, 1971, and 1975, respectively. He
joined Central Research Laboratory, Hitachi, Ltd. in 1974. From 1980 to 1981,
he was a visiting scientist at the Computer Science Department of Carnegie
Mellon University. He has been doing research and development work on
character recognition, document understanding, knowledge-based document
retrieval, and full-text search of Japanese documents. From 1992 to 1993, he
was a Chief Engineer of the Software Development Center. Currently, he is a
Chief Researcher at the Central Research Laboratory. He has also been a
lecturer for the Centre for Informatics at Waseda University since 1985.
He is a member of IEEE, AAAI, ACM, Information Processing Society of Japan,
and Institute for Electronics, Information and Communication Engineers, Japan.


Fourth Annual Symposium on
Document Analysis and Information Retrieval
University of Nevada, Las Vegas
April 24-26, 1995

Conference Registration Form

Name: ________________________________________________________________________

Title: ______________________________________________________________________

Company: ____________________________________________________________________

Address: ____________________________________________________________________

City: ______________________________ State: _______________ Zip: ____________

Telephone Number: (___________)_______________________________________________

E-mail Address:_______________________________________________________________

Registration Fees Pre-Reg Regular Amount
before 3/11/94 after 3/11/94

Conference Registration $400.00 $450.00 $____________
Includes dinner Monday, 4/24/95;
lunch Monday, 4/24/95
and lunch Tuesday, 4/25/95)

Dinner (Monday Dinner for Spouse/Companion) $ 10.00 $____________

Conference Proceedings (Extra Proceedings) $ 50.00 $____________
(One Proceedings is included as part of the
registration fee)

CD-ROM $100.00 $____________
(Conference Proceedings and Annual Report)

Make checks/money orders payable to: UNLV Board of Regents

Mail completed conference registration form and check/money order to:
Symposium Manager
Information Science Research Institute Telephone (702)895-4571
University of Nevada, Las Vegas Fax (702)895-1183
4505 Maryland Parkway Email This email address is being protected from spambots. You need JavaScript enabled to view it.
Box 454021
Las Vegas, NV 89154-4021 All checks/money orders must be in
U.S. Dollars and drawn on a U.S. Bank


Rooms Reserved Under the Name: SDAIR '95

Reservations received after March 23, 1995 will be accepted on a space
available basis only.

Please reserve accommodations for:

NAME: ________________________________________________________________________

HOME ADDRESS: ________________________________________________________________

CITY: _________________________ STATE: __________________ ZIP: _______________

HOME PHONE: _________________________________________________________________

COMPANY NAME: _______________________________________________________________

COMPANY ADDRESS: _____________________________________________________________

CITY: _________________________ STATE: __________________ ZIP: _______________

BUSINESS PHONE: _____________________________________________________________

SINGLE OCCUPANCY - $105.00 (+8% tax) DOUBLE OCCUPANCY - $105.00 (+8% tax)

WILL ARRIVE: _____________________________, 1995 TIME: ____________________

WILL DEPART: _____________________________, 1995 TIME: ____________________

Reservations must be received by March 23, 1995 - one night's deposit is
required to guarantee reservation. The following credit cards are accepted:

Enclosed is my one night's deposit payable by:

Check (Circle One) Credit Card

Mastercard _____ JCB _____ Visa _____ American Express _____

Carte Blanche _____ Discover _____ Diners Club _____

CREDIT CARD NUMBER: __________________________________________________________

EXPIRATION DATE: _____________________________________________________________

PLEASE PRINT NAME AS IT APPEARS ON CARD: _____________________________________

Room Reservations: (800)634-6906 Fax: (702)733-4676

Or mail your reservation to: Sheraton Desert Inn
3145 Las Vegas Blvd. South
Las Vegas, NV 89109



I also received a mail from the computer vision lab at the university
of Maryland announcing the availability of some technical reports. I
include this here as one of these reports (the first in the list) at
least is directly related to TC10's activities.

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

First of all, let me apologize for the email message everyone received
the other day which said merely "subscribe". It was an error. I was
trying to verify the method of subscribing to this mailing list and
left out a crucial part of the email address -- instead of sending
mail to "This email address is being protected from spambots. You need JavaScript enabled to view it. ", I left out the word
"request", so my request to subscribe was distributed to the
recipients of the mailing list by mistake.

Sandy German

Effective early in 1994, our technical reports became available in
compressed postscript format via anonymous ftp.
The postscript files have been placed in subdirectories according to
their publication dates (not the dates when actually printed,
but the dates when the report number was assigned).
Currently these subdirectories are called CVL-Reports-1993,
CVL-Reports-1994, and CVL-Reports-1995.

Below are the directions for anonymous ftp access to our archive:

1. "ftp ftp.cfar.umd.edu"

2. login "anonymous"

3. password "your login" (e.g., This email address is being protected from spambots. You need JavaScript enabled to view it. )

4. "cd CfAR/TRs" (please match upper and lower case as indicated)

As mentioned above, you may need to move into one of the subdirectories
CVL-Reports-1994 or CVL-Reports-1995.

The contents of these directories are:

a. "directory_contents" (an ascii file containing abstracts and other
descriptive information---keywords, postscript file name, total number
of pages when output in hardcopy, etc.---about each report available in
the directory)

b. "INDEX" (an ascii file containing the name of the postscript file
and the title of the paper contained in that file)

c. postscript files (some of these files can be quite large---even
when compressed---as they include figures and/or images)

Note: occasionally the title page of a report must be formatted
separately from the body of the report. When this happens the title
page will be posted as a separate postscript file with "tpg" as
part of the file name.

Advice: set "bin" (Apparently it is advisable to use binary transfer
when downloading compressed files---which all of these files are)


These files can also be accessed via the World Wide Web at:

http: //www.cfar.umd.edu/ftp/TRs


Most recent reports placed in the CVL report archive.

file name(s): CfAR/TRs/CVL-Reports-1994/TR3386-Samet.ps.Z
report numbers: CAR-TR-747 CS-TR-3386
title: MAGELLAN: Map Acquisition of GEographic Labels by Legend ANalysis
author(s): Hanan Samet and Aya Soffer
date: December 1994
pages: 18
support: IRI-9017393 and NGT-30130
agency: NSF and NASA
keywords: Map recognition, Document analysis, Object recognition,
Geographic Information Systems
A system named MAGELLAN (Map Acquisition of GEographic Labels by Legend
ANalysis) is described that utilizes the symbolic knowledge found in the
legend of the map to drive geographic symbol (or label) recognition.
MAGELLAN's output serves as input to a geographic information system (GIS).
MAGELLAN first scans the geographic symbol layer(s) of the map. The legend
of the map is located and segmented. The geographic symbols (i.e., labels)
are identified, and their semantic meaning is attached to them. An initial
training set library is constructed based on this information. The training
set library is subsequently used to classify geographic symbols in input
maps using statistical pattern recognition. User interaction is required at
first to assist in constructing the training set library to account for
variability in the symbols. The library is built dynamically by entering
only instances that add information to it. The library is stored in an
appropriate spatial data structure, and a highly efficient nearest neighbor
finding algorithm is used to search it. MAGELLAN then proceeds to identify
the geographic symbols in the input maps automatically. MAGELLAN can be
fine-tuned by the user to suit specific needs. An experimental study was
conducted on a large set of data and recognition rates of over 93\% were

file name(s): CfAR/TRs/CVL-Reports-1994/TR3387-Lin.ps.Z
title: Global Activity Detection: Contextual Recognition of Regular
Object Configurations
author(s): C.L. Lin, P. Burlina, and R. Chellappa
date: December 1994
pages: 41
support: DACA76-92-C-0024
keywords: Representation, detection and recognition of object
configurations, exploitation of remotely sensed imagery, model-based
change detection and monitoring, global operators, Quick-Look
Detecting change and monitoring trends and activities of mobile
objects are critical applications in model-supported exploitation of
aerial and satellite imagery. Recent efforts have shown the advantages
brought about by the use of site models to solve change detection (CD)
problems. Most model-supported approaches to CD use local object
detectors and infer changes by direct comparisons with previously
acquired images. We present instead an approach relying on the use of
global operators.

More specifically, our approach is based on a systematic analysis of
the description and detection of {\em regular object
configurations}. The proposed global operators rely on the spectral
analysis of regular and repeated edge structures and on enforcing
frequency domain constraints derived from known geometrical models of
monitored objects. A method for learning acceptance/critical regions
is presented. This approach is applied to aerial image exploitation
and the detection of vehicle formations such as convoys, platoons, and
the global monitoring of regions of interest such as parking areas,
roads and open areas. Results of extensive testing of these modules on
real images are reported.

file name(s): CfAR/TRs/CVL-Reports-1994/TR3389-Weiss.ps.Z
title: Physics-like Invariants for Vision
author(s): Isaac Weiss
date: December 1994
pages: 20
support: F49620-92-J-0332 and DACA76-92-C-0009
agency: AFOSR and ARPA/CETEC
keywords: invariants, physics-based vision, shape from shading, symmetries
Unlike geometric invariants, the invariants described here concern the
physical processes that form images, involving shading, IR, radar, sonar,
etc. The image formed by such a process depends on many variables in
addition to the geometry, such as the characteristics of the lighting or
other incident radiation, the imaging system, etc. Most of these variables
are not known in advance, so the recovery of shape is difficult. The
problem could be greatly simplified if we could find invariants of the
situation, namely quantities that stay unchanged as some of the unknown
variables change. In this paper we apply known methods of mathematical
physics to finding invariants of physical imaging processes. These methods
take advantage of various symmetries, which can be part of a model-based
approach to recognition. As an example we use the shape from shading
problem, but the methods have a much wider applicability.

file name(s): CfAR/TRs/CVL-Reports-1994/TR3390-Etemad.ps.Z
title: Separability Based Multiscale Basis Selection
and Feature Extraction for Signal and Image Classification
author(s): Kamran Etemad and Rama Chellappa
date: December 1994
pages: 26
support: MDA 9049-3C-7217
agency: ARPA
keywords: Separability, Basis Selection, Wavelet Packets, Dimensionality
Reduction, Segmentation, Textures, Document Images, Radar Signatures
Algorithms for multiscale basis selection and feature extraction for
general pattern classification problems are suggested. The basis selection
algorithm is based on class separability measures rather than energy or
entropy. At each level the ``overall'' and ``accumulated'' class
separabilities obtained from the tree for a parent node and its children
are computed and compared. The decomposition of the node (or subband) is
performed if it provides larger combined separability. The suggested
feature extraction algorithm focuses on dimensionality reduction of a
multiscale feature space subject to maximum preservation of information
useful for classification. At each level of decomposition, an optimal
linear transform that preserves class separabilities and results in a
reduced dimensional feature space is obtained. Classification and feature
extraction is then performed at each scale and resulting ``soft decisions"
are integrated across scales. The suggested algorithms have been tested for
classification and segmentation of 1D radar signals and 2D texture and
document images. The same idea can be used for other tree structured local
bases, e.g.\ local trigonometric basis functions, and even for
non-orthogonal, redundant and composite basis dictionaries.

file name(s): CfAR/TRs/Tech_Rept_Abstracts_1201-1299.ps.Z
title: Technical Reports of the Computer Vision Laboratory Nos. 1201-1299
date: February 1995
pages: 40
keywords: bibliography
This document gives the authors, titles, report numbers, contract or
grant numbers, dates, and abstracts of all technical reports produced by
the Computer Vision Laboratory, Center for Automation Research,
University of Maryland, College Park, MD 20742-3275, during the period
October 1992 -- February 1995.
The document itself will be counted as No. 1300 in the series.


Here is another CFP for a conference where document image analysis in
general and map & drawing analysis in particular are mentioned as
appropriate topics:

ICSC '95 - Third International Computer Science Conference:
Image Analysis Applications and Computer Graphics
Hong Kong, December 11-13, 1995


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


Invited Speaker: Invited Speaker:
Thomas Huang (University of Illinois) John Lasseter (PIXAR)

Industrial Visual Inspection Natural Phenomenon Modeling
Vision for Robotic Applications Shading & Rendering Algorithms
3D or Color Imaging in Inspection Image Synthesis & Realism
Document Image Processing Computer Animation
Printed Character Recognition Graphics for CAD/CAM
Cursive Script Recognition Graphics in Education
Map & Drawing Analysis Graphics & the Arts
Biomedical Image Analysis Graphics in Simulation
Image Analysis for Multimedia Scientific Visualization
Remote Sensing Image Analysis Virtual Reality
Software & Novel Systems Graphics in Multimedia


Authors are requested to submit manuscripts (in English) of no more than
5000 words describing original results of their research work. Papers
should be double-spaced and printed with high-quality printers using
type size of 11-point or larger. Each copy of the paper should have a
cover page containing the title of the paper, names and addresses of the
authors, and an abstract of no more than 200 words. Unless specified
otherwise, correspondence will be directed to the first author listed.
Four hard copies of each paper should be sent with a cover letter to:

Professor Roland T. Chin
Department of Computer Science
Hong Kong University of Science and Technology
Clear Water Bay, Hong Kong
E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Fax: +852 2358-1477


Tutorial sessions will be held during the conference. Proposals for
tutorials on the state of the art of image analysis and computer graphics
should be sent to the Program Chair.


Paper/Tutorial Submission Deadline: April 17, 1995
Notification of Acceptance: July 17, 1995
Camera-Ready Copy Due: September 15, 1995

General Chair: Horace Ip (City University of Hong Kong)

Program Chair: Roland Chin (University of Wisconsin & Hong Kong University of
Science and Technology)

International Program Committee:

Graphics Program Co-Chair: Avi Naiman (Hong Kong University of Science and

John Amanatides (York University)
Tony Derose (University of Washington)
Eugene Fiume (University of Toronto)
Alain Fournier (University of British Columbia)
Martin Gobel (FhG-IGD)
Pat Hanrahan (Princeton University)
Tosiyasu Kunii (University of Aizu)
Eihachiro Nakamae (Hiroshima Prefectural University)
Qunsheng Peng (Zhejiang University)
Demetri Terzopoulos (University of Toronto)
Keith Waters (Digital Equip. Corp.)
Geoff Wyvill (University of Otago)
Michael Zyda (Naval Postgraduate School)

Image Analysis Program Co-Chair: T.C. Pong (University of Minnesota & Hong Kong
University of Science and Technology)

Jake Aggarwal (University of Texas)
Narendra Ahuja (University of Illinois)
Terry Caelli (Curtin University of Technology)
Francis Chin (University of Hong Kong)
Charles Dyer (University of Wisconsin)
Robert Haralick (University of Washington)
Thomas Huang (University of Illinois)
Anil Jain (Michigan State University)
Josef Kittler (University of Surrey)
Song De Ma (Inst. of Automation, Beijing)
Linda Shapiro (University of Washington)
Yoshiaki Shirai (Osaka University)
Ching Y. Suen (Concordia University)
Demetri Terzopoulos (University of Toronto)
Saburo Tsuji (Osaka University)
Andrew Wong (University of Waterloo)
Guang-You Xu (Tsinghua University)

Organising Committee:

L.W. Chan (Chinese University of Hong Kong)
Ronald Chung (Chinese University of Hong Kong)
John C.M. Lee (Hong Kong University of Science and Technology)
Andrew Layfield (City University of Hong Kong)
Karl Leung (Hong Kong Polytechnic University)
Andrew Luk (City University of Hong Kong)
Joseph Ng (Baptist University)
H.F. Ting (University of Hong Kong)
Tong Lee (Chinese University of Hong Kong)
H.T. Tsui (Chinese University of Hong Kong)

Organised and Sponsored by IEEE (HK) Computer Chapter.
In cooperation with IEEE (USA) Computer Chapter, ACM (HK), Hong Kong
Computer Society, Hong Kong Institute of Engineers, Hong Kong Society
for Multimedia and Image Computing, and International Association for
Pattern Recognition.

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


Finally, I include some announcements which may be of less interest to
our community, but as they may be related in one way or the other to
the activities of some of us at least, I though it may be useful to
give the information.

First, an announcement about a conference on statistical shape analysis.
>From: This email address is being protected from spambots. You need JavaScript enabled to view it. (Ian Dryden)



International Conference run by Statistics Department,
University of Leeds, UK,
(incorporating 15th Leeds Statistics Research Workshop and CoMIR)

April 5-7, 1995

Statistical shape analysis considers the study of geometrical
objects where translation, rotation and possibly scale information
can typically be ignored. The subject has received a great deal of
recent attention, following the pioneering work of D.G. Kendall and
F.L. Bookstein in the late 1970s/early 1980s. Although the
methodology is being widely applied there are many key issues that
need to be resolved. The conference aims to bring together workers
in the field, to discuss both theoretical and practical issues.

11.35 am David Kendall Shapes and Shape-spaces:
Cambridge, UK What do they look like?
1.30 pm John Gower Distance, Geometry and Shape
Open, UK
2.15 pm Colin Goodall Procrustes Methods in the Statistical
Penn State, USA Analysis of Shape Revisited
3.30 pm Huiling Le The Mean Shape and the
Nottingham, UK Shape of the Means
4.00 pm Ian Dryden Investigating Regularity in Spatial
Leeds, UK Point Patterns using Shape Analysis
4.30 pm Subhash Lele Euclidean Distance Matrix Analysis
John Hopkins, USA of Landmark Coordinate Data:
a Statistical Review
9.00 am Kanti Mardia Shape Advances and Future
Leeds, UK Perspectives
9.30 am Burzin Bhavnagri Construction of a Markov Process
Adekaide, Australia to Model a Process Arising in Vision
10.00 am Ilya Molchanov Abstract Landmarks and their Applications
Amsterdam, NL
11.00 am Chris Small Shape Metrics and Frobenius Norms
Waterloo, Canada
Five-minute talks by poster presenters with discussion
2.00 pm Peter Green Reversible Jump MCMC Computation
Bristol, UK and High Level Image Analysis
2.45 pm John Marchant A Compartmented Snake for
AFRC, UK Model-based Boundary Location
3.45 pm Tim Cootes Active Shape Models: Recent Advances
Manchester, UK
4.15 pm Jens Anderson 2-D Shape Recognition by
Copenhagan, Denmark Sequential Test Procedures
4.45 pm Adam Baumberg Spatio-temporal Shape Model
Leeds, UK
5.15 pm Mads Nielsen Bayesian Inference in
Copenhagan, Denmark Computer Vision: Binocular
Stereo from Axioms to Algorithm
9.30 am Fred Bookstein The Morphometric Synthesis
Michigan, USA
11.00 am F.James Rohlf Multivariate Analyses of Shape
New York, USA
11.30 am Paul Sampson Eigenshape Analysis of Left
Washington, USA Ventricular Shape from 2-D Contrast
Ventriculograms and from 3-D
Trans-esophageal Echocardiography
12.00 noon Bill Green A Simple Construction of
Michigan, USA Triangle Shape Space
2.00 pm John Kent Topics in Inference for
Leeds, UK Shape Analysis
2.45 pm OPEN DISCUSSION PANEL - Fred Bookstein, Wilfrid Kendall,
John Kent

TIMETABLE: Wednesday 10.30 a.m.(registration) - Friday 4.00 p.m.

REGISTRATION FEE : 110 pounds sterling (students 90 pounds)
The conference will be held in Fairbairn House, part of the
University of Leeds campus, and which offers conference and
associated facilities to a high standard.
Fees include conference proceedings, lunch and car parking.

CONFERENCE DINNER (Thursday):25 pounds sterling,
payable with registration fee.
Fees should be paid in advance to UNIVERSITY OF LEEDS.

Accommodation in nearby hotels at reasonable prices can be arranged
if required. Individuals must settle their own accommodation

Dr. Christine Gill, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Department of Statistics,
University of Leeds, tel: 0113 2335157
Leeds, LS2 9JT fax: 0113 2335102



REGISTRATION FORM Please detach and return to Dr C. A. Gill,
Department of Statistics, University of Leeds, Leeds LS2 9JT.

NAME :...............................Fee enclosed(pounds sterling)
ADDRESS :............................... 110 or 90 ....
............................... Dinner 25 ....
............................... TOTAL ....
Please tick for:-
conference dinner.... car park space....
accommodation on: Tuesday night.... Wednesday night....
Thursday night.... Friday night....
any other nights, please specify.............
at price (pounds sterling): <30.... 30-40.... 40-50.... 50+....


Then, the first announcement of the next European Conference on
Computer Vision

>From This email address is being protected from spambots. You need JavaScript enabled to view it. (A. H. Gee)


ECCV '96
Fourth European Conference on Computer Vision

14-18 April 1996
University of Cambridge, England

Following the highly successful conferences held in Antibes, Santa
Margherita Ligure and Stockholm, the Fourth European Conference on
Computer Vision will be held from 14-18 April 1996 in Cambridge,
England. The Conference is to be held under the auspices of the
European Vision Society (EVS) and the British Machine Vision
Association (BMVA).

The programme will consist of a single track of the highest quality,
previously unpublished, contributed papers, delivered either orally or
as a poster. Contributions are sought on new research on any aspect of
computer vision.

All reviewing will be performed double blind by a Programme Committee
of leading international researchers selected by the Conference Board
and, as usual for the European Conference, the proceedings will be
published by Springer-Verlag.

Deadline for submission of papers 1 October 1995
Notification of acceptance 10 December 1995
Deadline for camera-ready manuscripts 20 January 1996
Conference begins 14 April 1996

A call for papers will be issued shortly. For further details of the
call for papers, conference programme, workshop facilities,
registration and local information please contact the Conference

ECCV'96 Conference Contact
42 Devonshire Road
Cambridge CB1 2BL

tel: +44/0 1223 323437
fax: +44/0 1223 460396
e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Reduced registration fees will be available to students and to members
of the EVS and national societies affiliated to the EVS.

In addition, as at previous Conferences, it is planned to hold a small
number of specialised workshops on Friday 19 April, immediately
following the main meeting. Anyone interested in organizing a workshop
is invited to contact the Conference Chairman (This email address is being protected from spambots. You need JavaScript enabled to view it. ).

Conference Chair:
Prof. Bernard Buxton (University College London)

Local Arrangements Chair:
Dr. Roberto Cipolla (University of Cambridge)

Conference Board

N.Ayache INRIA, Sophia Antipolis
M.Brady University of Oxford
H.Burkhardt University Hamburg-Harburg
J.Crowley INPG, Grenoble
E.Dickmanns Universitat der Bundeswehr, Munich
J-O.Eklundh Royal Institute of Technology, Stockholm
O.D.Faugeras INRIA, Sophia Antipolis
G.Granlund Linkoping University
D.Hogg Leeds University
J.Koenderink Utrecht State University
J.Mayhew Sheffield University
R.Mohr INPG, Grenoble
H-H.Nagel IITB, Karlsruhe
B.Neumann Hamburg University
G.Sandini DIST, University of Genova
V.Torre University of Genova


Finally, the CFP for a workshop on vision for robots:

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

Call for Papers


Sunday August 6th, 1995

The Westin William Penn Hotel, Pittsburgh, PA 15219

in conjunction with International Conference on
Intelligent Robots and Systems (IROS-95),
Aug 7-9, 1995.

General Chairman: Katsushi Ikeuchi (Carnegie Mellon Univ.)
Program Chairman: Avi Kak (Purdue Univ.)

Program Committee:
Minoru Asada (Osaka Univ.)
Peter Allen (Columbia Univ.)
Ruzena Bajcsy (Univ. of Pennsylvania)
Akio Kosaka (Purdue Univ.)
Martial Hebert (Carnegie Mellon Univ.)
Seth Hutchinson (Univ. of Illinois, Urbana-Champaign)
Charles Thorpe (Carnegie Mellon Univ.)

It could be said that the proof of the pudding in building a computer vision
system is to demonstrate it on a robot. After all, in order to make a robot
do anything useful through the use of its vision sensors, the vision system
must work with some degree of competency and robustness, not to mention the
fact that in addition to scene interpretation the vision system must also
yield information for the pose calculations needed for subsequent robotic
manipulation. Over the last decade, a number of research groups have
actually demonstrated robotic vision systems for both the arm robots and the
mobile robots. For the case of arm robots, researchers have demonstrated
bin-picking of non-polyhedral objects using 3-D vision systems. And, for
the case of mobile robots, researchers have demonstrated navigation modules
using monocular and binocular vision. The aim of this workshop is to bring
together people who have some experience with the integration of vision
systems with robots, both the arm robots and the mobile robots. Discussions
at the workshop will focus on defining more sharply the current
state-of-the-art in the design of such systems. We will also try to
delineate the next frontier of experiments for this kind of research.

More specifically, in addition to reviewing the progress in the design of
integrated robotic vision systems, the workshop will address a host of
technical questions that appear highly relevant to the field. For example,
while for bin-picking applications we have had great success with 3-D
vision, nothing comparable can be said for 2-D vision. Can any lessons
learned from 3-D vision be applied to crack the problem of 2-D vision? For
the case of mobile robot in indoor environments, is it better to use
precompiled models of the environment, or should the robot construct such
models using its sensors? What are the geometry vs. topology tradeoffs for
the representation of models for mobile robot navigation in general? What
are the best algorithms for pose calculation? Where do we stand in bridging
the gap between the closed-loop systems for visual servoing and model-based
systems for scene recognition? What about the use of motion cues? How
successful have been the concepts of purposive vision?


Four copies of the full paper including figures and drawings (double-spaced,
not exceeding thirty pages) must be received by Mar 1, 1995 to the Program


Submission of papers: Mar 1, 1995
Acceptance notification: May 1, 1995
Submission of final camera-ready papers: Jun 1, 1995

Well, once again this mail has become quite long. I hope you found at
least some of the information useful for you.

Many greetings from 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