5. CONCLUSIONS 
  
A novel approach for building reconstruction using both 
vertical and oblique images captured by Pictometry's imaging 
system is presented in this paper. It uses a building model 
which consists of three basic parts, i.e. points, lines and facets 
and their relations and a bottom-up strategy to reconstruct 
building roofs. Both the Moravec and the Canny operators are 
used to extract point and edge features. 3D features are 
generated by feature-based image matching and are grouped by 
using topological relations between features to generate roof 
primitives — facets. The method has been tested on different 
buildings in different images and test results show that building 
roofs can be reconstructed correctly if roof lines (eaves, ridge 
and valley lines) can be extracted properly. 
6. REFERENCES 
Dang, T., Jamet, O. and Maitre, H., 1994. Applying perceptual 
grouping and surface models to the detection and stereo 
reconstruction of buildings in aerial imagery. In: IAPRS, Vol. 
XXX, pp. 165-172. 
Fua, P. and Hanson, A.J., 1991. An optimization framework for 
feature extraction. Machine Vision and Application. Volume 4, 
pp. 59-87. 
Henricsson, O., 1996. Analysis of image structures using color 
attributes and similarity relations. PhD Thesis, Mitteilung,59, 
Institut für Geodásie und Photogrammtrie, Zürich, Switzerland. 
Huertas, A. and Nevatia, R., 1988. Detecting buildings in aerial 
images. Computer Vision, Graphics and Image Processing, 
Volume 41, pp.131-153. 
Lang, F. and Fórstner, W., 1996. Surface reconstruction of 
man-made objcts using polymorphic mid-level features and 
generic scene knowledge. In: IAPRS, Volume XXXI, pp. 415- 
420. 
Lin, C., Huertas, A. and Nevatia, R., 1995. Detection of 
buildings from monocular images. In: A. Grün, O. Kübler, and 
P. Agouris (eds.), Automatic Extraction of Man-Made Objects 
from Aerial and Space Images, pp. 125-134. Birkhäuser Verlag, 
Basel. 
Mayer, H., 1999, Automatic object extraction from aerial 
imagery — a survey focusing on buildings. Computer Vision and 
Image Understanding, Volume 74, issue 2, pp. 138-149. 
Roux, M. and McKeown, D.M., 1994, Feature matching for 
building extraction from multiple views. In: proceeding of 
DARPA Image Understanding workshop, pp. 331-349. 
Taillandier, F. and Deriche, R., 2004. Automatic building 
reconstruction from aerial images: a generic Bayesian 
framework. In. IAPRS, Volume XXV, Part 3, pp. 343-348. 
Xiao, J., Gerke, M. and Vosselman, G., 2010. Automatic 
detection of buildings with rectangular flat roofs from multi- 
view oblique imagery. In: IAPRS, Vol. XXVIII, Part 3A, pp. 
251-256. 
  
se; 
an 
int 
ca 
co 
re