Self-evaluation for active vision by the geometric information criterion
We present a scheme for evaluating the “goodness” of camer++a motion for robust 3-D reconstruction by means of the geometric information criterion (geometric AIC). The evaluation does not require any knowledge about the environment, the device, and the image processing techniques by which the images are obtained, and we need not introduce any thresholds to be adjusted empirically.
KeywordsFeature Point Image Noise Camera Motion Image Processing Technique Active Vision
Unable to display preview. Download preview PDF.
- 1.H. Akaike, “A new look at the statistical model identification,” IEEE Trans. Automation Control, 19-6 (1974), 176–723.Google Scholar
- 3.K. Kanatani, Geometric Computation for Machine Vision, Oxford University Press, Oxford, 1993.Google Scholar
- 4.K. Kanatani, “Renormalization for motion analysis: Statistically optimal algorithm,” IEICE Trans. Inf, & Sys., E77-D-11 (1994), 1233–1239.Google Scholar
- 5.K. Kanatani, “Automatic singularity test for motion analysis by an information criterion,” Proc. 4th European Conference on Computer Vision, April, 1996, Cambridge, U.K., pp. 697–708.Google Scholar
- 6.K. Kanatani, Statistical Optimization for Geometric Computation: Theory and Practice, Elsevier Science, Amsterdam, 1996.Google Scholar
- 7.K. Kanatani, “Geometric information criterion for model selection,” Int. J. Comput. Vision, to appear.Google Scholar
- 8.K. Kanatani and S. Takeda, “3-D motion analysis of a planar surface by renormalization,” IEICE Trans. Inf. & Syst., E78-D-8 (1995), 1074–1079.Google Scholar
- 9.H. C. Longuet-Higgins, “The reconstruction of a plane surface from two perspective projections,” Proc. Roy. Soc. Lond., B-227 (1986), 399–410.Google Scholar