Initial Matching of Multiple-View Images by Affine Approximation of Relative Distortions
To match multiple views of a 3D scene, their relative geometric distortions have to be taken into account. We assume the disortions can be approximated by affine transformations. Images are matched by combining an exhaustive and directed unconstrained Hooke-Jeeves search for affine parameters, image pyramids being used to accelerate the search. The parameters found for several matches are statistically processed to relatively orient the images. Experiments with the RADIUS multiple-view images show a feasibility of this approach.
Keywordsmultiple-view stereo image matching affine geometry
- 2.Förstner, W.: Image matching. In: Haralick, R.M., Shapiro, L.G.: Computer and Robot Vision. Vol. 2. Addison-Wesley, Reading (1993) 289–378Google Scholar
- 4.Koch, R., van Gool, L. (eds.): 3D Structure from Multiple Images of Large-Scale Environments Lecture Notes in Computer Science, Vol. 1506. Springer-Verlag, Berlin Heidelberg New York (1998)Google Scholar
- 8.RADIUS model board imagery and groundtruth (CD-ROM). Intelligent Machines Lab., University of Washington. Seattle (1996)Google Scholar
- 9.Shapiro, L.S.: Affine Analysis of Image Sequences. Cambridge University Press (1995)Google Scholar
- 10.Zhang, J.Q., Gimel’farb, G.: On detecting points-of-interest for relative orientation of stereo images. In: D. Pairman, H. North (eds.) Proc. Image and Vision Computing New Zealand 1999. Univ. of Canterbury, Christchurch, New Zealand, 30th–31st August 1999. Landcare Research, Lincoln (1999) 51–66Google Scholar