Date: 08 Feb 93 14:03:51-PST
From: Vision-List moderator Phil Kahn <Vision-List-Request@TELEOS.COM>
Errors-to: Vision-List-Errors@TELEOS.COM
Reply-to: Vision-List@TELEOS.COM
Subject: VISION-LIST digest 12.7
To: Vision-List@TELEOS.COM

VISION-LIST Digest    Mon Feb 08 14:03:52 PDT 93     Volume 12 : Issue 7

 - ***** The Vision List has changed hosts to TELEOS.COM *****
 - Send submissions to Vision-List@TELEOS.COM
 - Vision List Digest available via COMP.AI.VISION newsgroup
 - If you don't have access to COMP.AI.VISION, request list 
   membership to Vision-List-Request@TELEOS.COM
 - Access Vision List Archives via anonymous ftp to FTP.TELEOS.COM

Today's Topics:

 Gestures archives...
 Wanted: Fingerprint Recognition, Wavelets, Markov Random Fields
 Help: a question of GA
 Re: Frame grabber with adjustable gain
 laser radar systems
 looking for 2-D pattern matching software.
 ICCV93
 CFP: AAAI-93 Workshop on Reasoning About Function
 York Vision Conference
 CFP: Czech Pattern Recognition Workshop '93
 TR: Shape and Specularity from Color
 TR: Efficient Image Processing on RISC Workstations

----------------------------------------------------------------------

Date: Wed, 03 Feb 93 14:40:40 -0800
From: "Harpreet S. Sawhney" <sawhney@almaden.ibm.com>
Subject: Gestures archives...

Hullo Folks,

Is there a database of image sequences of hand gestures ?
Any pointers are very welcome.

Harpreet S. Sawhney         " From the Land of the Himalayas "
                            " And now close to the Sierras ! "
Research Staff Member
IBM Almaden Research Center, Dept. K54
650 Harry Road
San Jose, CA 95120

sawhney@almaden.ibm.com
408-927-1799 (work)
408-927-9048 (home)
408-927-4090 (FAX)

------------------------------

Date:         Wed, 03 Feb 93 10:01:55 MEZ
From: He-Ping Pan <ULM101@IBM.rhrz.uni-bonn.de>
Subject:      Wanted: Fingerprint Recognition, Wavelets, Markov Random Fields

I am looking for Literature and Software (public domain and/or commercial)
on:

  - Fingerprint Recognition
  - Wavelets for Texture Analysis and Other Purposes
  - Markov Random Fields for Texture Analysis

Your help will be highly appreciated.

Sincerely

Dr. He-Ping Pan
Institut fuer Photogrammetrie
Universitaet Bonn
Nussallee 15, 5300 Bonn 1, Germany
Email: ulm101@dbnrhrz1.bitnet

------------------------------

Date: 4 Feb 93 10:41:23 GMT
From: chen@kuri.ces.kyutech.ac.jp (Chen Ke)
Organization: Kyushu Institute of Technology, Iizuka, JAPAN.
Subject: Help: a question of GA

Hi, everybody!

I would appreciate it if anyone can give me the answer to the
following question:

  Could you give me an example about employing GA to solve 3-D object
recognition problem?

Answers are prefered with Email.

regards,
Ke Chen, Ph.D.
Masumi-Ishikawa Research Group
Dept. of Control Engineering and Science
Kyushu Institute of Technology
Iizuka, Fukuoka 820, Japan

Phone +81-948-29-7738(office)
  Fax +81-948-29-7709
Email chen@kuri.ces.kyutech.ac.jp

------------------------------

Date: Wed, 3 Feb 1993 18:52:20 -0500
From: rowe@lepomis.psych.upenn.edu (Mickey Rowe)
Subject: Re: Frame grabber with adjustable gain
Organization: University of Pennsylvania

>Date: Tue, 2 Feb 93 11:27:10 EST
>From: ubikkasa@mosaic.uncc.edu (Udayabhanu P Bikkasani)
>Subject: Frame grabber with adjustable gain
>
>I have two boards from the Imaging technology inc. They call it the
>PCvision Plus Framegrabber. It is IBMpc compatible and I am using it
>on a 386 for analysis and inspection of surface texture.

We also have a 386 with one of the PCvision Plus Framegrabber boards
in it.  However, we also have a series 150 image analysis system, also
made by Imaging Technology Incorporated.  If you're interested in this
company's products, I highly recommend that you try to go the more
expensive route and get the series 150.  Ours is equipped with an
analog-digital converter, one frame buffer board, and an arithmetic
logic unit board.  The difference in flexibility between the two
systems is considerable--in particular, if you want to write text or
objects on top of an image, you can do so non-destructively with the
series 150.  To the best of my ability to tell, that's not possible
with the PCVision Plus (ask me about dynamic overlays if you're
interested).

More germaine to the topic at hand, however, is that the series 150 is
noticeably more sensitive.  This device also has both gain and offset
control for the incoming analog signal, but settings range from 0-255
instead of 0-100.  If you have a problem with getting enough light to
your camera, you want the series 150.  With that setup I can see
images where I see nothing but black level noise in the PCVision
system.  

Although the series 150 is a seperate unit, we're using it in
conjunction with another 386 clone.

Mickey Rowe     (rowe@lepomis.psych.upenn.edu)

------------------------------

From: Jim Clark <jclark@cee.heriot-watt.ac.uk>
Subject: laser radar systems
Date: Fri, 5 Feb 93 15:02:54 GMT

 Hi, I'm interested in anybody who is using, has used, is thinking of using or   is developing a time of flight laser radar,
 be it phase, FM, chirp or direct measurement and or any companies and
 orginisations that can supply me with such a device.
 
 Thanks in advance
 
 Jim Clark
 Vision Research Group
 Heriot Watt University
 Edinburgh 
 
------------------------------

Date: Thu, 4 Feb 93 15:52:24 MET
From: Patrick Decowski <mpdecows@cs.ruu.nl>
Subject: looking for 2-D pattern matching software.

I'm looking for commercial software that will do pattern matching or
recognition. the scanned image has to be compared against a library of images
and then a best fit has to be selected. i'm interested in companies that
develop this software and more specifically in commercially available software
packages and on which system they run (crays excluded :->). as i don't read
this newsgroup frequently, direct e-mail would be prefered.
	thanks in advance,
Patrick Decowski,   University of Utrecht.
email: mpdecows@cs.ruu.nl	; van Lieflandlaan 60, 3571 AD, Utrecht.
tel #: ++31-30-732660		; 		the Netherlands.

------------------------------

Date: Mon, 8 Feb 93 17:38:14 MET
From: Kostas Daniilidis <kostas@ira.uka.de>
Subject: ICCV93

FOURTH INTERNATIONAL CONFERENCE ON COMPUTER VISION

ICCV '93 

ADVANCE PROGRAM 

Humboldt-Universitaet 
Berlin / Germany 

11-14 May 1993 

Jointly sponsored by 

IEEE Computer Society 
Gesellschaft fuer Informatik 

and co-sponsored by 

European Vision Society 
Fraunhofer-Gesellschaft zur Foerderung der 
angewandten Forschung e.V.  

For additional information contact:  
Deutsche Informatik-Akademie, Frau G. Trapp 
Ahrstr. 45, D-5300 Bonn 2/ Germany 
Phone: + 49-228-302165; Fax: + 49-228-37 86 90 

H.-H. Nagel, FhG-IITB, 
Fraunhoferstr. 1, D-7500 Karlsruhe 1/ Germany 
Phone: +49-721-6091-210; Fax: +49-721-6091-413 
e-mail: hhn@iitb.fhg.de 

General Chair 
Nagel, H.-H. (FhG-IITB und Universitaet Karlsruhe) 

Program Co-Chairs 
Nagel, H.-H. (FhG-IITB und Universitaet Karlsruhe) 
Huang, T.S. (Univ. Illinois, Urbana-Champaign / IL) 
Shirai, Y. (Osaka Univ., Osaka / Japan) 

Program Committee 
Ahuja, N. (Univ. Illinois, Urbana-Champaign / IL) 
Ayache, N. (INRIA Rocquencourt / France) 
Bajcsy, R. (U. Pennsylvania, Philadelphia / PA) 
Ballard, D. H. (U. of Rochester, Rochester / NY ) 
Blake, A. (Oxford University, Oxford / UK) 
Brady, M. (Oxford University, Oxford / UK) 
Brooks, M. J. (Univ. Adelaide, Adelaide / SA) 
Buxton, B. (GEC Hirst / UK) 
Eklundh, J.-O. (KTL Stockholm / Sweden) 
Faugeras, O. (INRIA Sophia Antipolis / France) 
Granlund, G. (Linkoping U., Linkoping / Sweden) 
Hasegawa, J.I. (Chukyo Univ., Toyota / Japan) 
Kanade, T. (Carnegie-Mellon U., Pittsburgh / PA) 
Koenderink, J.J. (Utrecht U., Utrecht / Netherlands) 
Kuno, Y. (Toshiba Corp., Kawasaki / Japan) 
Malik, J. (Univ. Calif., Berkeley / CA) 
Matsuyama, T. (Okayama U., Okayama / Japan) 
Mohr, R. (LIFIA Grenoble / France) 
Mundy, J. (GE Corpor. Research, Schenectady/NY) 
Ohta, Y. (Tsukuba Univ., Tsukuba / Japan) 
Pentland, A. (MIT Media Lab., Cambridge / MA) 
Rosenfeld, A. (U. of Maryland, College Park / MD) 
Sandini, G. (Univ. Genova, Genova / Italy) 
Shakunaga, T. (NTT Lab., Tokyo / Japan) 
Sugihara, K. ( Tokyo Univ., Tokyo / Japan) 
Terzopoulos, D. (Univ. Toronto, Toronto / ON) 
Tomita, F. (ETL, Tsukuba / Japan) 
Torre, V. (Univ. Genova, Genova / Italy) 
Tsotsos, J.K. (Univ. Toronto, Toronto / ON) 
Tsuji, S. (Osaka Univ., Osaka/Japan) 
Ullman, S. (Weizmann Institute, Rehovoth / Israel) 
Yang, H.S. (KAIST, Daejon / Korea) 
Zucker, S.W. (McGill Univ., Montreal / Quebec) 


ADVANCE PROGRAM

Monday, May 10, 1993 

16:00 - 20:00   Start of Registration 
                Foyer of the Auditorium Maximum,
                Humboldt-Universitaet,
                Berlin-Mitte, Unter den Linden 6 

18:30 - 22:00   Pre-Conference Gettogether 
                Opernpalais, Unter den Linden 5, 
                Berlin-Mitte (just opposite of the Humboldt-Universitaet) 



Tuesday, May 11, 1993 

8:00 - 9:00     Registration 
                Foyer of the Auditorium Maximum, Humboldt-Universitaet Berlin


9:00 - 9:10     Opening and Welcome 
                Auditorium Maximum, Humboldt-Universitaet Berlin 

OPTICAL FLOW / MOTION DETECTION 

9:10 - 9:40     A Generalized Brightness Change Model for Computing 
                Optical Flow 
                S. Negahdaripour, C.-H. Yu 

9:40 - 10:00    Robust Computation of Optical Flow in a Multi-Scale 
                Differential Framework 
                J. Weber, J. Malik 

10:00 - 10:30   Motion Detection Robust to Perturbations:  a Statistical 
                Regularization and Temporal Integration Framework
                J.M. Letang, V. Rebuffel, P. Bouthemy 

10:30 - 11:00   Coffee Break 

PERCEPTUAL ORGANIZATION 

11:00 - 11:30   A Computational Model of Neural Contour Processing:  Figure-
                Ground Segregation and Illusory Contours 
                F. Heitger, R. von der Heydt 

11:30 - 11:50   Active Exploration: Knowing When We're Wrong 
                P. Whaite, F.P. Ferrie 


11:50 - 12:10   Looking for Trouble: Using Causal Semantics to Direct Focus of
                Attention 
                L. Birnbaum, M. Brand, P. Cooper 

12:10 - 12:30   Ridge-Detection for the Perceptual Organization Without Edges
                J.B. Subiana-Vilanova, K.K. Sung 


12:30 - 14:30   Lunch Break 

TRACKING IN 2-D AND 3-D 

14:30 - 15:00   Affine-invariant Contour Tracking with Automatic Control 
                of Spatiotemporal Scale 
                A. Blake, R. Curwen, A. Zisserman 

15:00 - 15:20   Tracking Foveated Corner Clusters Using Affine Structure 
                I.D. Reid, D.W. Murray 

15:20 - 15:40   Experiments with Monocular Visual Tracking and Environment 
                Modeling 
                O. Silven, T. Repo 

15:40 - 16:00   Tracking Non-Rigid Objects in Complex Scenes 
                D.P. Huttenlocher, J.J. Noh, W.J. Rucklidge 


16:00 - 16:30   Coffee Break 

RECOGNITION AND LEARNING 

16:30 - 17:00   A Spherical Representation for the Recognition of Curved 
                Objects 
                H. Delingette, M. Hebert, K. Ikeuchi 

17:00 - 17:20   Fast and Robust 3D Recognition by Alignment 
                T.D. Alter, W.E.L. Grimson 

17:20 - 17:40   Learning Recognition and Segmentation of 3-D Objects from 2-D
                Images 
                J.J. Weng, N. Ahuja, T.S. Huang 

17:40- 18:00    Vision-Based Construction of CAD  Models from Range Images 
                X. Chen, F. Schmitt 

19:00 -         Conference Reception 



Wednesday, May 12, 1993 

ILLUMINATION AND COLOR 

9:00 - 9:30     Diffuse Shading, Visibility Fields, and the Geometry
                of Ambient Light 
                M.S. Langer, S.W. Zucker 

9:30 - 9:50     Estimation of the Light Source Distribution 
                and its Use in Shape Recovery from Stereo and Shading
                D. Hougen, N. Ahuja 

9:50 - 10:10    Recovering Reflectance and Illumination in a World of Painted 
                Polyhedra 
                P. Sinha, E. Adelson 

10:10 - 10:30   Diagonal Transforms Suffice for Color Constancy 
                G.D. Finlayson, M.S. Drew, B.V. Funt 

10:30 - 11:00   Coffee Break 


PICTURE DOMAIN APPROACHES 

11:00 - 11:30   A General Method for Image Fusion in Computer Vision 
                P.J. Burt, R.J. Kolczynski 

11:30 - 11:50   An Extension of Marr's Signature Based Edge Classification and
                Other Methods for Determination of Diffuseness and Height of
                Edges, as Well as Line Width 
                W. Zhang, F. Bergholm 

11:50 - 12:10   Segmentation and 2D Motion Estimation by Region Fragments 
                M. Etoh, Y. Shirai 

12:10 - 12:30   Energy-Minimizing Segmentation of an Image 
                A. Ackah-Miezan, A. Gagalowicz 

12:30 - 14:00   Lunch Break 

14:00 - 16:00 POSTER SESSION I 
( Ordered alphabetically according to first author) 

Optical Flow from 1D Correlation: Application to a Simple
Time-To-Crash Detector
N. Ancona, T. Poggio 

Mathematical Morphology: The Hamilton-Jacobi Connection LEMS- 108 A. Arehart,
L. Vincent, B.E. Kimia 

Visual Echo Analysis 
E. Bandari, J.J. Little 

Localization and Positioning Using Combinations of Model Views 
R. Basri, E. Rivlin 

A Framework for the Robust Estimation of Optical Flow 
M.J. Black, P. Anandan 

A Design for a Visual Motion Transducer 
A. Blake, G. Hamid, L. Tarassenko 

Building and Using Flexible Models Incorporating Grey-Level Information 
T.F. Cootes, C.J. Taylor, A. Lanitis, D.C. Cooper, J. Graham 

A Quantitative Methodology for Analyzing the Performance of Detection 
Algorithms 
T. Kanungo, M.Y. Jaisimha, J. Palmer, R M. Haralick 

Multiscale Markov Random Field Models for Parallel Image Classification 
Z. Kato, M. Berthod, J. Zerubia 

A Perceptually Plausible Model for Global Symmetry Detection 
F. Labonte, Y. Shapira, P. Cohen 

3D Object Recognition by Indexing Structural Invariants from Multiple Views 
R. Mohan, D. Weinshall, R.R. Sarukkai 

Silhouette-Based Object Recognition Through Curvature Scale Space 
F. Mokhtarian, H. Murase 

Fast Segmentation, Tracking and Analysis of Deformable Objects 
Ch. Nastar, N. Ayache 

Reflectance Ratio: A Photometric Invariant for Object Recognition 
S.K. Nayar, R.D. Bolle 

Contextual Feature Similarities for Model-Based Object Recognition 
D. Noll, M. Schwarzinger, W. von Seelen 

A Linear Complexity Procedure for Labelling Line Drawings of
Polyhedral Scenes Using Vanishing Points 
P. Parodi, V. Torre

Learning Object Recognition Models from Images 
A.R. Pope, D.G. Lowe 

Recognition of Object Classes from Range Data 
I.D. Reid, J.M. Brady 

A Finite-Element Framework for Correspondence and Description 
S. Sclaroff, A. Pentland 

Incremental Image Sequence Enhancement with Implicit Motion Compensation 
A. Singh 

Recognizing Mice, Vegetables and Hand Printed Characters Based on
Implicit Polynomials, Invariants and Bayesian Methods
J. Subrahmonia, D. Keren, D.B. Cooper 

Understanding Noise: The Crucial Role of Motion Error in Scene Reconstruction 
J.I. Thomas, A. Hanson, J. Oliensis

The Reciprocal-Wedge Transform for Space-Variant Sensing 
F. Tong, Z.-N. Li 

Automatic Feature Point Extraction and Tracking in Image Sequences for
Unknown Camera Motion 
Q. Zheng, R. Chellappa

Eliciting Qualitative Structure from Image Curve Deformations 
A. Zisserman, A. Blake, C.A. Rothwell, L.J. Van Gool 

16:00 - 16:30 Coffee Break 

SCENE PARAMETER ESTIMATION 

16:30 - 17:00 Exploiting the Generic View Assumption to Estimate Scene
              Parameters 
              W. T. Freeman 

17:00 - 17:20 Combining Stereo and Motion Analysis for Direct Estimation 
              of Scene Structure 
              K.J. Hanna, N.E. Okamoto 

17:20 - 17:40 Motion Segmentation and Local Structure 
              D. Sinclair 

17:40 - 18:00 Robust Structure from Motion Using Motion Parallax 
              R. Cipolla, Y. Okamoto, Y. Kuno 



Thursday, May 13, 1993 

ACTIVE VISION 

9:00 - 9:30   Robust Vergence with Concurrent Detection of Occlusion 
              and Specular Highlights 
              W.-S. Ching, P.-S Toh, K.-L. Chan,  M.-H. Er 


9.30 - 9:50   Head Centered Orientation Strategies in Animate Vision 
              E. Grosso, D.H. Ballard 

9:50 - 10:10  Reactions to Peripheral Image Motion Using a Head-Eye Platform 
              D.W. Murray, P.F. McLauchlan, I.D. Reid, P.M. Sharkey 

10:10 - 10:30 Dynamic Fixation 
              K. Pahlavan, T. Uhlin, J.-0. Eklundh 

10:30 - 11:00 Coffee Break 



STEREO / CALIBRATION 

11:00 - 11:30 Stereo Matching, Reconstruction and Refinement 
              of 3D Curves Using Deformable Contours 
              B. Bascle, R. Deriche 

11:30 - 11:50 A Binocular Stereo Algorithm for Reconstructing Sloping, Creased,
              and Broken Surfaces in the Presence of Half-Occlusion 
              N. Belhumeur 

11:50 - 12:10 A Complete Two-plane Camera Calibration Method and Experimental
              Comparisons 
              G.-Q. Wei, S.D. Ma 

12:10 - 12:30 Direct Estimation of Multiple Disparities of Transparent Multiple
              Surfaces in Binocular Stereo 
              M. Shizawa 

12:30 - 14:00 Lunch Break 

14:00 - 16:00 POSTER SESSION II

A System for Automatic Vectorization and Interpretation of
Map-Drawings 
S.  Ablameyko, V. Bereishik, O. Frantskevich, M. Homenko, A. Melnik,
O. Okun, N. Paramonova

Texture Discrimination by Local Generalized Symmetry 
Y. Bonneh, D. Reisfeld, Y. Yeshurun 

Euclidean Constraints for Uncalibrated Reconstruction 
B. Boufama, R. Mohr, F. Veillon 

A Front End System for Dynamic Vision 
P.J. Burt, P. Anandan, K.J. Hanna, G. van der Wal 

Projectively Invariant Decomposition and Recognition of Planar Shapes 
S. Carlsson 

Calibrating an Active Stereo Head to Object Centered Reference Frames 
J.L. Crowley, P. Bobet, C. Schmid 

Recognizing Algebraic Surfaces from their Outlines 
D. Forsyth 

Relative Depth from Vergence Micromovements 
A. Francisco 

Distance Accumulation and Planar Curvature 
J. H. Han, T. Poston 

Using Hyperquadratics for Shape Recovery from Range Data 
S. Han, D.B. Goldgof, K.W. Bowyer 

Interpretation of Natural Scenes Using Multi-Parameter Default Models
and Qualitative Constraints 
M. Hild, Y. Shirai

Cooperation of Visual Guided Behaviors 
J. Kosecka, R. Bajcsy 

A Note on Existence and Uniqueness in Shape from Shading 
R. Kozera 

Minimum Description Length Based 2-D Shape Description 
M. Li 

A Finite Element Model for 3D Shape Reconstruction and Nonrigid Motion Tracking
T. McInerney, D. Terzopoulos 

Surface Discontinuities in Range Images 
T. Pajdla, V. Hlavac 

Accurate Line Parameters from an Optimising Hough Transform for
Vanishing Point Detection 
P.L. Palmer, M. Petrou, J. Kittler

Optimal Estimation of Object Pose from a Single Perspective View 
T.Q. Phong, R. Horaud, A. Yassine, P.D. Tao 

Relative 3D Positioning and 3D Convex Hull Computation from a Weakly
Calibrated Stereo Pair 
L. Robert, O.D. Faugeras

Robust Line-Based Pose Enumeration from a Single Image 
T. Shakunaga 

Occam Algorithms for Computing Visual Motion 
H. Shvaytser 

2-D Digital Curve Analysis: a Regularity Measure 
B. Vasselle, G. Giraudon 

A Robust Active Contour Model with Insensitive Parameters 
G. Xu, E. Segawa, S. Tsuji 

Robust 3D-3D Pose Estimation 
X. Zhuang, Y. Huang 

16:00 - 16:30 Coffee Break 

STRUCTURE ESTIMATION 

16:30 - 17:00 Extracting Projective Structure from 
              Single Perspective Views of 3D Point Sets 
              C.A. Rothwell, D.A. Forsyth, A. Zisserman, J. L. Mundy 

17:00 - 17:20 Projective Structure from two Uncalibrated Images:  Structure
              from Motion and Recognition 
              A. Shashua 

17:20 - 17:40 Computation of Ego-Motion and Structure 
              from Visual and Inertial Sensors Using the Vertical Cue 
              T. Vieville, P.E. Dos Santos Facao, E. Clergue


17:40 - 18:00 Renormalization for Unbiased Estimation 
              K. Kanatani 

19:00-        CONFERENCE BANQUET 



Friday, May 14,1993 

MACHINE VISION APPROACHES FOR ROAD VEHICLES 

9:00 - 9:20   An All-Transputer Visual Autobahn-Autopilot/Copilot 
              E.D. Dickmanns, R. Behringer, C. Bruedigam, 
              D. Dickmanns, F. Thomanek, V. v. Holt 

9:20 - 9:40   Active and Intelligent Sensing of Road Obstacles: 
              Applications to the European Eureka-PROMETHEUS Project 
              M. Xie, L. Trassoudaine, J. Alion, M. Thonnat, J. Gallice 

9:40 - 10:00  Multiple Knowledge Sources and Evidential Reasoning for Shape
              Recognition 
              B. Besserer, B. Ulmer 

10:00 - 10:20 Quantitative Analysis of the Viewpoint Consistency Constraint in
              Model-Based Vision 
              L. Du, G.D. Sullivan, K.D. Baker 

10:20 - 10:40 Coffee Break 

ESTIMATION OF DISPLACEMENT, POSE, AND MODEL PARAMETERS 

10:40 - 11:00 Integration of Quantitative and Qualitative Techniques for
              Deformable Model Fitting from Orthographic, Perspective, and
              Stereo Projections 
              D. Metaxas, S.J. Dickinson 

11:00 - 11:20 Large Deformable Splines, Crest Lines and Matching 
              A. Gueziec 

11:20 - 11:40 An Improved Algorithm for Algebraic Curve and Surface Fitting 
              G. Taubin 

11:40 - 12:00 Probabilistic Relaxation for Matching Problems in Computer Vision
              J. Kittler, W. Christmas, M. Petrou 

12:00 - 13:00 Lunch Break 

SHAPE FROM SHADING / TEXTURE / MOTION 

13:00 - 13:20 Linear and Incremental Acquisition of Invariant 
              Shape Models from Image Sequences 
              D. Weinshall, C. Tomasi 

13:20 - 13:40 Shape from Texture from a Multi-Scale Perspective 
              T. Lindeberg, J. Garding 

13:40 - 14:10 A General Algorithm for Shape from Shading 
              J. Oliensis, P. Dupuis 

14:10 - 14:30 Coffee Break 


INFERRING 3-D CHARACTERISTICS 

14:30 - 15:00 Egomotion Analysis Based on the Frenet-Serret Motion Model 
              Z. Duric, A. Rosenfeld, L.S. Davis 

15:00 - 15:30 The Critical Sets of Lines for Camera Displacement Estimation: 
              A Mixed Euclidean-Projective and Constructive Approach 
              N. Navab, O.D. Faugeras, T. Vieville 

15:30 - 16:00 Grasping Visual Symmetry 
              A. Blake, M. Taylor, A. Cox 

16:00         END OF ICCV'93 



         GENERAL INFORMATION 

LOCATION 

The conference will be held in the Humboldt-Universitaet Berlin, Unter
den Linden 6, right in the center of Berlin: stepping out of the
Humboldt-Universitaet and turning right, you can see the Brandenburger
Tor - it is within walking distance.  Moreover, the
Humboldt-Universitaet is in the immediate neighborhood of
world-reknown cultural attractions.

REGISTRATION 

Please register by sending the enclosed registration form by mail or
fax to the ICCV'93 Conference Office. Confirmation will be sent after
receipt of the registration form and accompanying payment. The
registration fee covers the conference handout including one copy of
the proceedings, coffee and cookies during all coffee breaks (no
guarantee of an unlimited supply of all goodies!), the reception on
May 11, 1993 as well as the conference banquet on May 13, 1993.
Payment must be enclosed with your registration by one of the
alternatives mentioned on the registration form.

CANCELLATIONS 

In case of cancellation prior to April 11th, 1993, a processing charge
of 25 % will be withheld from the fee to be refunded. No refunds will
be possible for cancellations arriving on or later than April 11th, 1993.

HOTEL ACCOMMODATION 

As it is difficult to find hotel accommodation in Berlin, a number of
rooms have been reserved in conveniently located hotels up until March
15th, 1993 (see list below). Each type of accommodation will be
available in limited quantities only, to be booked on a first-come,
first-served basis (Keyword: ICCV '93). We cannot guarantee
accommodation after March 15th, 1993. In case none of the reserved
hotel rooms should be available any more, please contact the

Tourist Office 
Europa Center, Tauentzienstr. 9 
D-1000 Berlin 30 / Germany 

Phone +49-30-2123-2583/-2302 
Fax   +49-30-2123-2520 

Please make your (binding) hotel reservation directly with the hotel
(not via the conference office ) by fax; give credit card number and
expiration date for deposit in case of late arrival. The payment of
accommodation has to be settled by each participant.

LIST OF HOTELS in which rooms have been reserved for participants of ICCV '93.
The price given covers bed and breakfast per night for a 1-bed room (one person
only, lower price) or for a 2-bed room (two persons, higher price). The number
of rooms reserved in each category is given in parentheses.  

* Hotel Charlottenhof, 
  Charlottenstr. 52, 
  1080 Berlin / Germany, 
  Phone +49-30-23806-0;  
  Fax +49-30-23806-100 
  DM 156,- (40) / DM 190,- (5) 

* Hotel Unter den Linden, 
  Unter den Linden 14, 
  1080 Berlin / Germany, 
  Phone +49-30-23811-317;  
  Fax +49-30-23811-100;  
  DM 170,- (70) / DM 250,- (20) 

* Forum Hotel Berlin, 
  Am Alexanderplatz, 
  Postfach 47, 
  1020 Berlin / Germany, 
  Phone +49-30-2389-0;  
  Fax +49-30-2389-4305;  
  DM 200,- (200) / DM 260,- (60) 

* Hilton Hotel Krone, 
  Kronenstr. 48 
  1080 Berlin / Germany, 
  Phone +49-30-2382-4255;  
  Fax +49-30-2382-4324;  
  DM 259,- (50) / DM 283,- (10) 

* Hilton Hotel, 
  Mohrenstr. 30, 
  1080 Berlin / Germany, 
  Phone + 49-30-2382-4255;  
  Fax +49-30-2382-4324;  
  DM 384,- (10) / DM 463,- (10) 




ICCV '93 Registration Form      (PLEASE PRINT)

Surname:.............................................................

First Name, Initial:.................................................

GI or IEEE member number: ...........................................

Affiliation:.........................................................

Address:.............................................................

City / State / ZIP: .................................................

Country: ............................................................

Daytime Phone: +.....................................................

Fax:+................................................................


Fee for registrations arriving at the registration office prior to
April 1, 1993

GI or IEEE members:        DM 450,- 
non-members:               DM 600,- 

Fee for registrations arriving at the registration office on or after
April 1, 1993

GI or IEEE members:        DM 540,- 
non-members:               DM 690,- 

All payments must be made in 
Deutsche Mark (DM) 

O    Visa         O   MasterCard 

credit card number: .....................  

expiration date: ........................  


O  by cheque payable to "Gesellschaft fuer 
   Informatik, for ICCV '93" (ensure your 
   name appears clearly on the cheque) 

O  by money transfer to "Gesellschaft fuer Informatik" 
   account number: 41970
   name of the bank: Sparkasse Bonn 
   keyword: ICCV '93, name of participant 
   bank address: D-5300 Bonn 1/ Germany 
   bank code: 380 500 00 

 Date: .................              Signature: ........................

Please return by mail or fax to 

Deutsche Informatik-Akademie GmbH 
att.: Mrs. G. Trapp / ICCV '93 
Wissenschaftszentrum 
Ahrstr. 45 
D-5300 Bonn 2/ Germany 
Fax:   + 49-228-37 86 90 
Phone: + 49-228-302 165 



------------------------------

Date: Sun, 7 Feb 1993 22:53:44 GMT
From: amruth@cs.buffalo.edu (Amruth Kumar)
Organization: State University of New York at Buffalo/Comp Sci
Subject: CFP: AAAI-93 Workshop on Reasoning About Function

                    Call for Participation
                    ----------------------
                     AAAI-93 Workshop on
                   Reasoning About Function
                   ========================
                   Washington D.C., July '93

Description of Workshop: 

The explicit representation and use of the function (purpose) of an object,
either as intended by its designer or as interpreted by its user, is
emerging as a focal point of problem solving in fields as diverse as
Device Invention, Redesign, Diagnosis, Explanation Generation and
Automatic Debugging.  Explicit treatment of function has proven to be
very useful because of its potential to organize and provide access to
causal knowledge of the object (eg., focuses on missing causality
during redesign), because of the improved resolution it brings
to the reasoning process (eg., discriminates among suspects
during diagnosis) and because of its utility in addressing the scaling
problem. To date, function-based knowledge has been successfully
applied in several domains, including Software Engineering, Human
Physiology, and various fields of engineering such as Electrical,
Aerospace and Chemical Engineering.
 
The objectives of this workshop are: 1) to examine current techniques
used to represent and reason about function; 2) to present a forum to
develop a shared framework for reasoning about function, and 3) to
identify the trends and future directions for this emerging field.

Topics

The workshop will focus on the following issues:

1) Terminology: theoretical analyses yielding an appropriate
vocabulary for function; disambiguating function from other terms such
as behavior and teleology: what are the interactions and dependencies
between them? How do form, use, experience, etc. relate to function?

2) Representation: What are the issues in acquisition of function
knowledge? What are the ontological bases of an adequate function
representation: processes, states, flows, parameters? How can
representational primitives be chosen in a given ontology? What impact
does the intended use of a function model have on its construction?
How is knowledge of function related to behavioral, structural and
heuristic knowledge?

3) Reasoning: What are the sorts of tasks/domains for which Functional
Reasoning is particularly well suited?  What are the AI processes
involved in exploiting function knowledge during reasoning? How can
reasoning about function, behavior, teleology, etc. be integrated into
a coherent system?

4) Applications: An examination of implemented systems, i.e.,
representation and reasoning techniques used, evaluation of the system
and lessons learned from the experience.

Format of Workshop:

The workshop will address the work of participants in the form of
moderated presentations and discussions. Where appropriate,
participants will be invited to display posters describing their work.
Separate sessions will be devoted to terminology, representation
issues, reasoning and application issues. The workshop will conclude
with a summarizing panel discussion.

Attendance: 

Participation is by invitation only, and will be limited to
approximately 35 people. See Submission Requirements below for the
criteria to be invited.

Submission Requirements: 

Those who wish to attend the workshop should submit four copies of a
1-2 page research summary including a list of relevant publications,
regular and email address (where possible) and phone and FAX number.
Those who wish to present their work at the workshop should submit
four copies of a short paper (6-8 pages) in addition to the research
summary. If the work has been published or submitted for consideration
elsewhere, please specify the journal or conference.  Note that,
unpublished work is preferred for presentation. Electronic submissions
will not be accepted. All submissions will be reviewed by the Workshop
Committee.

In order to facilitate interaction among participants, you are
requested to specify the following with regard to your work: topic
(terminology/representation/reasoning/application), field
(redesign/diagnosis/explanation generation etc.) and domain
(electrical/chemical/physiology etc.). In submitted papers, you are
urged to either define or relate to existing literature, all
terminology used.  Also, please include a two-line description of the
main contribution of your paper.

Submission Deadline: March 12, 1993

Notification Date: April 2, 1993

Final date for camera-ready copies to organizers: April 30,1993

Submit to:

Amruth N. Kumar
(Reasoning About Function Workshop)
 
226 Bell Hall
Department of Computer Science
SUNY Buffalo, NY 14260
Ph: (716) 645 2193 Fax: (716) 645 3464
(amruth@cs.buffalo.edu)

Workshop Committee:

Dean Allemang 
Swiss Federal Institute of Technology (EPFL)
(allemang@lia.di.epfl.ch)

David Franke 
Trilogy Development Group
(franke@trilogy.com)

Jack Hodges
San Francisco State University
(hodges@huckleberry.sfsu.edu)

Amruth N. Kumar (Coordinator)
SUNY Buffalo
(amruth@cs.buffalo.edu)

James K. McDowell
Michigan State University
(mcdowelj@pleiades.cps.msu.edu)

Jon Sticklen
Michigan State University
(sticklen@cps.msu.edu)

Shambhu J. Upadhyaya
SUNY Buffalo

------------------------------

Date: 	Thu, 4 Feb 1993 14:12:00 -0500
From: "Michael Jenkin" <jenkin@cs.yorku.ca>
Subject: York Vision Conference

International Conference And Nato Advanced Research Workshop on 
Binocular Stereopsis and Optic Flow
 
June 22-June 26, 1993
York University
Toronto, Canada
 
 
REGISTRATION:
 
NAME:   ______________________________________________________________
 
ADDRESS ____________________________________________________________
         
AFFILIATION ________________________________________________________

            ____________________________________________________________
 
Tel. No.  & E-mail  _______________________________________________________
 
I enclose registration fee*
 
                                $__________


FACULTY & OTHERS $60.00 CDN  ($52. U.S.)
 
STUDENTS $30.00 CDN  ($26. U.S.)



                                $__________
 


I enclose $35 Can. ($30. U.S.) for the banquet on June 24st.



                                $__________
 
Total enclosed


                                $_________
 

DEDUCT $5. FROM THE REGISTRATION FEE IF PAID BEFORE  MAY 15th.
 
Accommodations:
I will require the following accommodation (no deposit required):
In the colleges of York University at $40. single per night with shared
facilities or $60. double with shared facilities. Prices include all taxes.
 
                                College single             College Double**
 
Tuesday June 22nd                _____                          _____
Wednesday June 23th              _____                          _____
Thursday June 24th               _____                          _____
Friday June 25th                 _____                          _____
Saturday June 26th               _____                          _____
 
Hotel reservations at the nearby Hotel Strata - $54 per night for single room or
$57 per night for a double room. Prices include all taxes and a continental
breakfast with hotel accommodations only.  You will be responsible for your own
hotel accommodations.
For hotel reservations please call Chris or Irv at (416) 665-3500 or 
Fax (416) 665-0807.
 
POSTERS
Facilities are available for Posters.
Please submit a title and abstract (about 100 words) before March 31st.
 
Please return this form to:
 
Teresa Manini
103 Farquharson Building
York University
North York, Ontario, M3J 1P3
Canada
 
Tel (416) 736 5659   Fax (416) 736 5857   E-mail manini@hpl.ists.ca
 
*Registration and banquet fees should be in Canadian Dollars (or US equivalent)
 by Check or Money payable to York University.
 
** LIMITED NUMBER OF DOUBLE ROOMS, FIRST COME FIRST SERVE BASIS.
 
This conference is sponsored by NATO, NSERC, YORK UNIVERSITY AND THE INSTITUTE
 FOR SPACE AND TERRESTRIAL SCIENCE


------------------------------

Date:         3 Feb 93 08:02:57 CET
From: "Vasek Hlavac"  <HLAVAC@vision.felk.cvut.cs>
Organization: Czech Technical University
Subject:      CFP: Czech Pattern Recognition Workshop '93

            November 4th - 6th, 1993
        Prenet, Sumava mountains, Czech Republic

         First announcement and call for papers

MAIN TOPICS - Theory and applications in:
-     Pattern Recognition        -     Image Processing
-     Computer Vision            -     Speech Recognition

AIM OF THE WORKSHOP
      The event is aimed at providing Czech Pattern Recognition
Society (part of the Czech Society for Cybernetics and
Informatics, Member of International Association for Pattern
Recognition) community and foreign researchers with an
opportunity to present scientific results, exchange ideas, share
details of their own work with others, and establish personal
contacts.

GENERAL INFORMATION
      The workshop is organized by the Czech Pattern Recognition
Society. The scientific programme will comprise two invited
lectures, oral and poster presentation. Papers will be published
in the proceedings and handed to participants at the workshop.
      The working language of the workshop is English.

VENUE
      The workshop will be held at the chalet Prenet owned by the
Faculty of Electrical Engineering, Czech Technical University.
The chalet is situated in the beautiful scenery on the ridge of
the Sumava mountains at 1000 metres above sea level 20 km to the
south from the town Klatovy.
      The amount of participants is limited to forty. First the
intention is to have workshop tiny and friendly. Second reason
is the limited capacity of the chalet. There will be a bus
departing from Praha on Thursday November 4th in the morning and
coming back to Praha on Saturday November 6th in the early
evening.

CONFERENCE SECRETARIAT
Vaclav Hlavac, Czech Technical University,
Faculty of Electrical Engineering,
Department of Control Engineering,
121 35 Praha 2, Karlovo namesti 13, Czech Republic
FAX: +42 2 290159   Phone: +42 2 295664
E-mail: hlavac@vision.felk.cvut.cs

SUBMISSION OF PAPERS Authors wishing to present paper should
submit two copies of a paper in English (max. 5 A4 pages). The
submitted papers will be reviewed by two reviewers. Authors will
receive a notification of acceptance of their papers. Accepted
papers should be submitted in camera ready form.

SCIENTIFIC AND ORGANIZING COMMITTEE
- Stanislava Simberova (Co-chairman, Mrs), Astronomical
  Institute, Czech Academy of Sciences, Observatory Ondrejov
- Vaclav Hlavac (Co-chairman), Czech Technical University, Praha
- Ivan Bajla, Institute of Measurement, Slovak Academy
  of Sciences, Bratislava
- Jan Flusser, Institute of Information Theory and Automation,
  Czech Academy of Sciences, Praha
- Michal Haindl, Centrum voor Wiskunde en Informatica,
  Amsterdam, The Nederlands
- Josef Psutka, West-Bohemian University, Plzen

IMPORTANT DATES
January 20, 1993  First Announcement
May 20, 1993      Deadline for submission of 3 copies of papers
                  and/or Registration form
July 15, 1993     Notification to authors, issue of Final
                  Programme and logistic information to all
                  participants
Sept. 15, 1993    Deadline to receive camera ready papers from
                  authors
November 4, 1993  Opening of the workshop

WORKSHOP FEES     (includes accommodation, meal, organization
                  fees, proceedings)

if paid before September 15, 1993
700 Czech Crowns  for members of Czech Pattern Recognition Society
150 DM            for other participants

late payment
800 Czech Crowns  for members of Czech Pattern Recognition Society
170 DM            for other participants

=================================================================
Czech Pattern Recognition Workshop, APPLICATION FORM
(Please fill in and return by E-mail, FAX or ordinary mail now. )

Name/Title:
Affiliation:
Address:
Phone:              FAX:                     E-mail:
Tentative title of my contribution:

I intend to participate at the workshop (please tick)      yes/no
I prefer oral         yes/no     poster presentation       yes/no

------------------------------

Date: Thu, 4 Feb 1993 15:04:45 -0800
From: Mark Drew <mark@cs.sfu.ca>
Subject: TR: Shape and Specularity from Color

The following Tech Report is available for anonymous ftp
from cs.sfu.ca.  The file is in compressed .ps format, so use
> ftp cs.sfu.ca
> anonymous
> cd pub/techreports
> bin
> get CSSLCCR93-01.ps.Z
> quit
> uncompress CSSLCCR93-01.ps
The color postscript files look better on a screen than printed,
unsurprisingly.

\begin{Abstract} 
  For  a  Lambertian  surface,  it is possible to recover a dense
representation of shape in the form of surface  normals  at  each
visible  point by using a single color image of the surface.  The
positions, strengths, and spectral content  of  illuminants  need
not  be  known  a priori [1].  This capability is contingent upon
the surface being illuminated  by  a  set  of  lights  that  vary
spectrally   with   orientation   from  the  surface,  and  is  a
generalization of the photometric stereo  method.    This  "shape
from   color"   method  is  thus  restricted  to  one  particular
reflectance model, and works for a single surface at a time only.
The  Lambertian model means that the surface normal is related in
a linear way to the  RGB  color  observed  for  each  pixel;  the
linearity  holds only if the lights are linearly independent both
in space and in color space.  Surface normals are recovered up to
an  overall  orthogonal transformation.  Linearity means that the
Gaussian sphere is transformed into an ellipsoid in color  space,
and  one  can easily solve for the ellipsoid using least squares.
Once one  introduces  another  reflectance  model,  however,  the
linearity  and  hence  the  solvability of the model breaks down.
Nevertheless,  if  the  linearity-breaking   mechanism   is   the
addition of specularities to a base Lambertian reflectance it may
still be  possible  to  recover  shape  if  many  pixels  produce
reflected  light according to the Lambertian model.  In that case
one may treat  specular  highlights  as  outliers  on  top  of  a
basically  ellipsoidal shell in RGB space.  Several robust models
exist for discounting such outliers.  Here we use a least  median
of  squares  method.    Once  one  has  identified the Lambertian
ellipsoid, outliers, now identified as specular  highlights,  may
be  omitted  from  an  analysis of shape.  Thus one both recovers
shape and detects specularities at the same time.
1.    M.S. Drew, "Shape from Color", Tech. report  CSS/LCCR  TR
      92-07,   Simon   Fraser   University  School  of  Computing
      Science,1992.
\end{Abstract}

Shape and Specularity from Color
@TechReport(  DREW.SPECSHAPECOLOR.TR.92,
  Title       = {Shape and Specularity from Color},
  Author      = {M.S. Drew},
  Institution = {Simon Fraser University School of Computing Science},
  Number      = {CSS/LCCR TR 93-01},
  Year        = {1993}
           )


------------------------------

Date: Mon, 8 Feb 1993 04:20:24 GMT
From: len@retina.mpce.mq.edu.au (Len Hamey)
Organization: School of MPCE, Macquarie University, Australia.
Subject: TR: Efficient Image Processing on RISC Workstations

The following technical report is available for anonymous
FTP from ftp.mpce.mq.edu.au (137.111.216.12) in
/pub/comp/techreports/920113.hamey.ps.Z
The abstract is available as /pub/comp/techreports/920113.hamey.abs

	Efficient Image Processing on RISC Workstations

		   Leonard G. C. Hamey
		   Computing Department
	 	   Macquarie University
		   NSW 2109 AUSTRALIA
		   len@mpce.mq.edu.au

	   Macquarie Computing Reports 92-0113C

		   December 1992

Abstract

The use of Reduced Instruction Set Computers for scientific applications is
common today. Advances in processor design and compiler technology make it
possible to perform large-scale computations on RISC workstations. The RISC
design provides simple instructions that operate at high speed and compiler
optimisations employ the machine's capabilities to perform naively programmed
operations with reasonable efficiency. However, there remain opportunities for
improvement in execution performance by optimising the design of the high-
level language program, and, conversely, there are also traps for the unwary
programmer. Partial unrolling of loops and employing scalar variables to assist
the compiler in the use of registers are two simple techniques that can yield
performance improvements of up to 50% depending upon the compiler and the
architecture. Unexpected performance penalties as high as 500% may be paid for
cache conflicts, where two large data arrays are contending for the same cache
locations. This report discusses how to improve the performance of critical
code sections in scientific programs, and how to detect and avoid cache
conflict performance penalties.

47 pages.


C language source code used in this paper may be obtained by
anonymous FTP from ftp.mpce.mq.edu.au in the directory
/pub/src/920113.hamey.

To fetch and print a copy of the file, use the following procedure:

	% ftp ftp.mpce.mq.edu.au
	Name: anonymous
	Password: <user>@<site.domain>
	ftp> cd /pub/comp/techreports
	ftp> binary
	ftp> get 920113.hamey.ps.Z
	ftp> quit
	% uncompress 920113.hamey.ps.Z
	% lpr -P<postscript-printer> 920113.hamey.ps

If your site does not support the Unix uncompress program, or you
have trouble fetching compressed files, our server can uncompress
the file for you.  Simply fetch 920113.hamey.ps instead of 920113.hamey.ps.Z.

	Leonard G. C. Hamey (Len)	len@mpce.mq.edu.au
	Lecturer in Computing		(02)805-8978
	Macquarie University
	NSW 2109 AUSTRALIA

------------------------------

End of VISION-LIST digest 12.7
************************
