> tv
-
AS
nar, but
lanar to
within a
connect
a
v.
fers to a
3, a pair
(). Note
D») of
vo lines
) have a
jundary
s of one
in order
>s Is the
nputing
er those
is more
ons are
back to
have to
ted yet.
5) and
her 110
Y and 5
e effect
ess than
'secting
Stephan Heuel
6 CONCLUSIONS
This paper presents a procedure for 3D grouping by combination of topological and geometrical reasoning. It shows the
advantage of using topological reasoning to efficiently reduce the number of geometric checks to be performed while
feature grouping. The topological reasoning is based on a close interaction of 2D and 3D information to exploit the
redundancy of the observed FAG's in image space. While subsequently checking geometrical relations we use rigorous
statistical testing which allowed to use the same control parameters for all examples and thus supports the generality of
the approach. the experiments with real data demonstrated the feasibility of the approach. Future work is directed toward
integrating single points, but more important, lines or wings ((Baillard et al., 1999)) and possibly regions or meshes in
order to improve completeness of the reconstructions. We also will investigate the impact of integrating range data, either
from stereo or from laser scanners, (Brunn, 2000). ?
REFERENCES
Baillard, C., Schmid, C., Zisserman, A. and Fitzgibbon, A., 1999. Automatic line matching and 3d reconstruction of
buildings from multiple views. In: ISPRSGIS99, pp. 69-80.
Brooks, R., 1987. Model-based three-dimensional interpretation of two-dimensional images. In: M. Fischler and
O. Firschein (eds), Readings in Computer Vision, Morgan Kaufmann, pp. 360—370.
Brunn, A., 2000. A step towards semantic-based building reconstuction using markov-random-fields. In: Proceedings of
the XIXth ISPRS Congress, Vol. XXXIII, Part B, ISPRS, Amsterdam.
Clowes, M., 1971. On seeing things. Artificial Intelligence 2, pp. 79-116.
Dold, A., 1972. Lectures on Algebraic Topology. Springer, Berlin.
Forstner, W., 1994. A Framework for Low Level Feature Extraction. In: J. O. Eklundh (ed.), Computer Vision - ECCV
94, Vol. II, LNCS, Vol. 802, Springer, pp. 383—394.
Forstner, W., 2000a. New orientation procedures. In: Proceedings of the XIXth ISPRS Congress, Vol. XXXIII, ISPRS,
Amsterdam.
Förstner, W., 2000b. Repräsentation und Prüfung von Beziehungen unsicherer geometrischer Elemente im Raum. In:
J. Albertz (ed.), Photogrammetrie und Fernerkundung — Neue Sensoren/Neue Anwendungen, Vol. 8, DGPF, Berlin.
Frere, D., Hendrickx, M., Vandekerckhove, J., Moons, T. and van Gool, L. ., 1997. On the reconstruction of urban house
roofs from aerial images. In: G. A., E. Baltsavias and O. Henricsson (eds), Automatic Extraction of Man-Made Objects
from Aerial and Space Images (II), pp. 87-96.
Fuchs, C., 1998. Extraktion polymorpher Bildstrukturen und ihre topologische und geometrische Gruppierung. DGK,
Bayer. Akademie der Wissenschaften, Reihe C, Heft 502.
Henricsson, O., 1996. Analysis of Image Structures using Color Attributes and Similarity Relations. PhD thesis, Swiss
Federal Institute of Technology, ETH Zürich, No. 11663.
Herman, M. and Kanade, T., 1986. The 3d mosaic scene understanding system. In: A. Pentland (ed.), From Pixels to
Predicates, Norwood N.J. Ablex Publ. Co.
Kanatani, K., 1995. Statistical Optimization for Geometric Computation: Theory and Practice. 2 edn, AI Laboratory,
Gunma University, Japan.
Lang, F. and Fórstner, W., 1996. Surface reconstruction of man-made objects using polymorphic mid-level features and
generic scene knowledge. Zeitschrift für Photogrammetrie und Fernerkundung 6, pp. 193-201.
Rothwell, C., Mundy, J. and Hoffman, B., 1996. Representing objects using topology. In: J. Ponce, A. Zisserman and
M. Hebert (eds), ECCV '96, International Workshop, Lecture Notes in Computer Science, Springer-Verlag, pp. 80-108.
Roux, M. and McKeown, D. M., 1994. Feature matching for building extraction from multiple views. In: ARPA IU
Workshop '94, Vol. I, Monterey, California, Morgan Kaufman, ISBN 1-55860-338-7. 13-16 Nov.
Sarkar, S. and Boyer, K., 1993. Perceptual organization in computer vision: A review and a proposal for a classificatory
structure. Transaction on Systems, Man, and Cybernetics 23, pp. 382—399.
Weber, M., 1998. Quadrik Through three Lines. personal communication, Bonn.
* Acknowledgements: The authors would like to thank Thomas H. Kolbe and Markus Zink for providing parts of the implementation.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000. 403
