ISPRS Commission III, Vol.34, Part 3A ,,Photogrammetric Computer Vision‘, Graz, 2002 
  
PREFACE 
It is an honor for our Institute to be invited to organize the quadrennial International Symposium of ISPRS 
Commission III, held in Graz, Austria, from 9 to 12 September 2002. After all, our Institute was created in the 
fall of 1992 and is currently celebrating its 10" birthday. It may easily have been considered too immature to 
organize an event with a nearly 100-year history. Thank you, ISPRS-leadership, for this vote of confidence! 
This book is Part A of the proceedings of that symposium and presents those papers that had been submitted to 
and were accepted by a double blind peer review of full papers, not abstracts. “Double blind” means that the 
reviewers did not know the identity of the authors, and the authors had no knowledge of the identity of the 
reviewers. 
ISPRS Commission III addresses the theory and algorithms of photogrammetry and remote sensing. Among the 
seven ISPRS Commissions it may therefore be the most academically oriented, with less in governmental or 
commercial interests. It is therefore only logical that it would be Commission III to introduce for the first time in 
ISPRS-history the process of double-blind-peer-reviews of full papers to assemble a program for one of its 
quadrennial symposia. In the past these reviews were based on abstracts only, and were not “blind”. 
The excellent work of the 9 Working Groups of the Commission led to the submission of 90 full papers for 
review. Of these, 67 could be accepted for inclusion in this Part A. Not all of the accepted papers could be 
presented orally at the Symposium since it had room for 45 oral presentations. The review involved 21 persons 
as listed on the next page. I need to thank all of them for their free labor and dedication. They reflect the 
Working Group leadership of the Commission. 
In addition to the full papers, authors were also invited to propose contributions in the ISPRS-tradition based on 
abstracts. Many authors preferred to stay with this approach, and we had therefore to find a way to accommodate 
both the new full papers as well as the traditional abstracts. Papers developed from the abstracts were not 
reviewed and are presented in a separate Part B of these proceedings. These contributions were not considered 
for oral presentations. 
For almost 100 years, photogrammetry had a monopoly on the development of technologies and theories of 
photographic measurement and interpretation. However, since the late 1960's, computer science has evolved 
very rapidly into a separate academic discipline. It is very recent that advances in computer technology have 
spawned the emergence of computer vision. A separate vision-community evolved hesitantly from the mid- 
1980s as a new field within computer science. The current year 2002 demonstrates that computer vision has 
enormously grown, has diversified into more than 10 sub-communities and has moved center stage within 
computer science. It now is a formidable and very broad alternative home to scientific work with images. We 
believe that there is room for both a photogrammetric as well as a computer science approach to “seeing by 
computer”. At the intersection is “photogrammetric computer vision”, and to emphasize this idea, we chose this 
term as an alias for this year’s symposium. 
The look over the “photogrammetric fence” is also visible in the selection of the three keynote speakers of the 
symposium. Luc van Gool, Giulio Sandini and Gerd Hirzinger have their homes in various segments of 
computer vision. We invited them to present an overview of their work. Instead of creating a new write-up of 
their widely published work, we include pointers to their work in Part A of these proceedings. I want to thank 
these keynote speakers for their contribution to the conference! 
Recent ISPRS-symposia have become multi-track events with many parallel sessions. Inspired by the standards 
of international conferences on computer vision, we wanted to revert back to the single-track organizational 
format that was customary in early ISPRS symposia. We hope that the work in selecting a program of oral 
presentations will pay off in a rewarding experience for the symposium attendees. 
Franz Leberl 
2002 Conference Chair 
A - iii