Affine Reconstruction Based on Parallel Plane and Infinity Point

Abstract:

Affine reconstruction is to restore the affine shape of the object. Generally, there are two ways of achieving, one is to determine the plane at infinity, another is to determine the plane homography. Using the homography which had determined the plane at infinity achieve affine reconstruction. In this paper, firstly give out the homography of infinity plane and the algorithm of affine reconstruction, then proved: if the scene contains a set of parallel planes and a infinity point, the homography of infinity plane can be obtained and affine reconstruction can be linearly got in the scene. Computer simulation and real experiments show that the linear affine reconstruction algorithm is correct, and the approach has a good precision.

• Periodical: Advanced Materials Research (Volumes 255-260)
• Pages: 2272-2275
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.255-260.2272
• Online since: May 2011
• Authors: Yue Zhao, Xiao Hua Hu
• Keywords: Affine Reconstruction, Computer Vision, Infinite Homography, Point at Infinity

References

[1] Fauferas O. What can been seen in three dimensions with an uncalibrated stereo. (In Proc ECCV92, Santa Margherita Ligure, Springer-Verlag, Italy 1992), pp.563-578.

DOI: 10.1007/3-540-55426-2_61

[2] Pollefeys M., Gool Van L., Osterlinck A. The modulus constraint: A New Constraint for self-calibration, In: Proceedings of International conference on Pattern Recognition, Vienna Austria, Vol. 31-42 (1996), pp.349-353.

DOI: 10.1109/icpr.1996.546047

[3] Pollefeys M. Koch R. Gool VL. Self-calibration and metric reconstruction in spite of varying and unknown internal camera parameters. In: ( Proceedings of International conference on Computer Vision. Bombay, 1998), pp.90-95.

DOI: 10.1109/iccv.1998.710705

[4] Wu FC. HU ZY. A new theory and algorithm of linear camera self-calibration. Chinese Journal of Computers, 24(11) ( 2001), pp.1121-1135.

[5] Sun-Fengmei, Hu-Zhangyi. Some properties about Homography matrix constraint on the intrinsic parameters. Journal of Computer-Aided and computer Graphics. 19(5) ( 2007), p.647 – 650.

[6] Ma-Songde, Zhang-Zhenyou. Computer Vision. Computing theory and Algorithm basics. (Science Press, BeiJing 1998), pp.89-92.

[7] Hartley R, Zisserman A. Multiple View Geometry In Computer Vision. (Cambridge Cambridge University Press, U K 2000).

[8] Zhao-Weimin, Liang-Dong. Stratified reconstruction of three-dimensional objects based on two images. Computer Engineering and Applications. 36 (2003), pp.78-80.

[9] Wu-Fucao, Hu-Zhangyi. Determine homography matrix of infinite plane linearly and camera calibration . Acta automatica sinica. 28(4) ( 2002), pp.487-496.

[10] Sun-Fengmei, Wu-Fucao, Hu-Zhangyi. Determine homography matrix of infinite plane based on projection of parallel planes. Journal of Software. 14(5) (2003), pp.935-946.

[11] Wu-Fucao ect. Mathematical Methods of Computer Vision. (Science Press, BeiJing 2008), pp.63-75, pp.104-105.