CIRA 2003 XIX th International Symposium, 30 September - 04 October, 2003, Antalya, Turkey
479
Because of the processing limitations caused by the hard- and
software which have been mentioned above, the 3D model of
the whole cenotaph structure will remain relatively coarse as
compared to the relief models. The 2 mm sampling as well as
the poorer accuracy of the 3D points result in a good geometric
model for the whole structure; detail resolution and neighbor
hood accuracy are not satisfactory, however. The procedures to
be carried out are much the same as described for the high
resolution model (section 4.1). A combination of all scanned
and meshed information is not possible presently. There is little
doubt that this can be accomplished in the near future.
Fig. 7: Detail of figure 6. Scale about 1:1
5. RESULTS AND CONCLUSIONS
Results of the documentation project are presented in figures 4
through 8. The full value of the data can only be judged when
evaluated in 3D on a computer monitor, however. 3D scanning
yields results that have not been possible in the past for objects
with extensive and complicated 3D surfaces as in the case of the
cenotaph and the associated reliefs. Photogrammetric matching
methods to achieve a digital object model do not work in this
case as the white marble reliefs do not show enough texture.
Orthophotos or line drawings from stereo pairs can be useful for
some purposes but do not contain the information for the
creation of a complete virtual (or real) model.
The processing of the laser scanner data is very time-
consuming, however, when high model quality is aimed at.
Presently, it also suffers from many restrictions. Even with the
latest computers and software products, certain processing steps
are only possible when the number of meshed triangles is less
than some millions. If the development in hard- and software
continues as rapidly as in the past, these problems should be
overcome in a few years. Even a combination of laser scanner
(geometric) and photogrammetric (texture) data may become
available for such large and complicated objects. Since all the
original cenotaph data are archived, improved results may be
created in the future.
6. ACKNOWLEDEMENTS
We want to thank Westcam Datentechnik GmbH for the fruitful
cooperation during the measurement process. Linsinger Ver-
messung, a private surveying company specializing in cultural
heritage documentation, did an excellent job performing the
photogrammetric documentation of the monument. Our
acknowledgement refers - last not least - to the local authorities
of Tyrol (Land Tirol, Landesbaudirektion) for the financial
support and the continuous assistance solving the administrative
problems.
Fig. 8: Same detail as in figure 7, but seen from different angle.
7. REFERENCES
Geomagic, 2003; Homepage http:/www.geomagic.com
GOM, 2003; Homepage www.gom.com
Hanke, Klaus, 2003: Dokumentation des Grabmals Kaiser
Maximilian I. in der Innsbrucker Hofkirche. In: Chesi, Weinold
(Hrsg.) “12. Internationale Geodätische Woche Obergurgl
2003”. Wichmann-Verlag. ISBN 3-87907-401-1
i3mainz, 2003:
3D scanning web site: http://scanning.fh-mainz.de
Marbs, Andreas, 2002: Experiences with laser scanning at
i3mainz. Proc. of the CIPA WG6 Int. Workshop on scanning for
cultural heritage recording.
http://www.isprs.org/commission5/workshop/
Mensi, 2003: Homepage www.mensi.com
