Title :
Affine structure from translational motion with varying and unknown focal length
Author_Institution :
Sch. of Technol. & Soc., Malmo Univ., Sweden
Abstract :
In this paper a method for obtaining affine structure from an image sequence taken by a translating camera with varying and unknown focal length is presented. A general geometric constraint, expressed using the camera matrices, is derived and this constraint is used in a least squares solution of the problem. The proposed algorithm extends the previous result of affine structure recovery from an image sequence with a translating camera with constant intrinsic parameters to the case with a camera with varying and unknown focal length and otherwise known intrinsic parameters. The noise sensitivity of the method is evaluated in simulated experiments.
Keywords :
computer vision; image reconstruction; image sequences; least squares approximations; matrix algebra; affine structure recovery; camera matrices; geometric constraint; image sequence; intrinsic parameters; least squares solution; noise sensitivity; translational motion; unknown focal length; Application software; Cameras; Computational modeling; Computer simulation; Computer vision; Image reconstruction; Image sequences; Layout; Least squares methods; Vehicle crash testing;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334022