DocumentCode :
456930
Title :
Towards Robust Voxel-Coloring: Handling Camera Calibration Errors and Partial Emptiness of Surface Voxels
Author :
Anwar, Zeeshan ; Ferrie, Frank
Author_Institution :
Center for Intelligent Machines, McGill Univ., Montreal, Que.
Volume :
1
fYear :
0
fDate :
0-0 0
Firstpage :
98
Lastpage :
102
Abstract :
In this paper, we present two new methods to reduce the effects of camera calibration errors and partial emptiness of surface voxels on voxel-coloring. Both of these sources of error introduce outlier pixels in voxel projections in the input images and thus result in over-carving of the reconstructed 3D scene. The existing methods to handle these errors are either insufficient or too complex. Our proposed methods are simple and can be incorporated into existing voxel-coloring algorithms easily. Our experimental results show that the methods proposed in this paper have the ability to improve the results of existing algorithms
Keywords :
Gaussian processes; calibration; cameras; computer vision; image colour analysis; image reconstruction; stereo image processing; 3D scene reconstruction; adaptive Gaussian averaging; camera calibration errors; outlier pixels; partial emptiness; surface voxels; voxel coloring; voxel projections; Additive noise; Calibration; Colored noise; Image reconstruction; Layout; Machine intelligence; Pixel; Robustness; Smart cameras; Surface reconstruction;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 2006. ICPR 2006. 18th International Conference on
Conference_Location :
Hong Kong
ISSN :
1051-4651
Print_ISBN :
0-7695-2521-0
Type :
conf
DOI :
10.1109/ICPR.2006.1129
Filename :
1698842
Link To Document :
بازگشت