DocumentCode :
3256191
Title :
3D Modeling from Multiple Views with Integrated Registration and Data Fusion
Author :
Boyer, Alain ; Curtis, Phillip ; Payeur, Pierre
Author_Institution :
Sch. of Inf. Technol. & Eng., Univ. of Ottawa, Ottawa, ON, Canada
fYear :
2009
fDate :
25-27 May 2009
Firstpage :
252
Lastpage :
259
Abstract :
This paper presents an integrated modeling system capable of generating coloured three dimensional representations of a scene observed from multiple viewpoints. Emphasis is given to the integration of the components and to the algorithms used for acquisition, registration and final surface mapping. First, a sensor operating with structured light is used to acquire 3D and colour data of a scene from multiple views. Second, a frequency-domain based registration algorithm computes the transformation between pairs of views from the raw measurements and without a priori knowledge on the transformation parameters. Finally, the registered views are merged together and refined to create a rich 3D model of the objects. Real world modeling examples are presented and analyzed to validate the operation of the proposed integrated modeling system.
Keywords :
frequency-domain analysis; image colour analysis; image fusion; image registration; image representation; solid modelling; 3D modeling system; coloured three dimensional representation; data fusion; frequency-domain-based registration algorithm; integrated image registration; surface mapping; Calibration; Clouds; Computer vision; Costs; Frequency domain analysis; Frequency estimation; Image reconstruction; Iterative closest point algorithm; Layout; Robot sensing systems; data fusion; frequency registration; object modeling; scene reconstruction; structured light;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Robot Vision, 2009. CRV '09. Canadian Conference on
Conference_Location :
Kelowna, BC
Print_ISBN :
978-0-7695-3651-4
Type :
conf
DOI :
10.1109/CRV.2009.39
Filename :
5230510
Link To Document :
بازگشت