DocumentCode :
2705480
Title :
An automatic method for texture extraction and mapping in 3d building reconstruction
Author :
Cheng Hui ; Zhang Jianpei
Author_Institution :
Coll. of Comput. Sci. & Technol., Haerbin Eng. Univ., Haerbin, China
fYear :
2012
fDate :
6-8 June 2012
Firstpage :
322
Lastpage :
325
Abstract :
A method for texture extraction and mapping form 3d building image is proposed. by the images, the 3d-building reconstruction is based on the plane. Texture images are then extracted by merging perspective corrected input images. A new technique for removing perspective distortion and for the estimation of the texture image color to compensate for the light source color is studied. The method avoids any intermediate 3D measurement. A texture mapping strategy based on surface particles is developed to adequately address photography related problems such as inhomogeneous lighting, highlights and occlusion. Reconstruction results are included to demonstrate the attained quality. Experiments show the effectiveness of the correction method using range intensity images.
Keywords :
CAD; feature extraction; image colour analysis; image reconstruction; image texture; photography; solid modelling; structural engineering computing; 3D building image; 3D building reconstruction; automatic method; correction method; highlights; inhomogeneous lighting; light source color; occlusion; perspective distortion; photography; range intensity image; surface particle; texture extraction; texture image color estimation; texture image extraction; texture mapping; Buildings; Computational modeling; Image color analysis; Image reconstruction; Lighting; Shape; Vectors; Color correction; Texture mapping; image rectification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information and Automation (ICIA), 2012 International Conference on
Conference_Location :
Shenyang
Print_ISBN :
978-1-4673-2238-6
Electronic_ISBN :
978-1-4673-2236-2
Type :
conf
DOI :
10.1109/ICInfA.2012.6246826
Filename :
6246826
Link To Document :
بازگشت