A method for buildings reconstruction

Author

Xue, Qiang ; Zhang, Zhi-qiang ; Sun, Ji-zhou

Author_Institution

Dept. of Comput. Sci. & Technol., Tianjin Univ., China

Volume

5

fYear

2003

fDate

2-5 Nov. 2003

Firstpage

2855

Abstract

To perform the image-based modeling (IBM) procedure, the parameters, such as camera focus and principal point, are always needed by user input or by complicated calculation. A new computer vision based IBM method for building reconstruction is presented in this paper. It reconstructs the building models by projective reconstruction at first, and then makes use of the surface features of buildings to complete the Euclidean reconstruction and optimizes the results. In view of the experiment results, the method recovers the 3-dimension information of the objects satisfactorily and has the advantages of simple operations and less additive information needed.

Keywords

building; computer graphics; computer vision; image reconstruction; 3 dimension information; Euclidean reconstruction; IBM; additive information; building surface features; buildings reconstruction; camera focus; computer vision; image based modeling; Buildings; Cameras; Focusing; Image reconstruction; Layout; Matrix decomposition; Reconstruction algorithms; Sun; Surface reconstruction; Surface texture;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2003 International Conference on

Print_ISBN

0-7803-8131-9

Type

conf

DOI

10.1109/ICMLC.2003.1260044

Filename

1260044