B. A. Maxwell and S. A. Shafer, "Segmentation and Interpretation of Multi-colored Objects with Highlights", Computer Vision and Image Understanding, Vol.77, No.1 January 2000, pp. 1-24. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Geometric MDL and Its Media Applications - Kanatani, Matsunaga (2000) (Correct)
....information criterion) 1] Another well known information criterion in statistics is Rissmen s MDL (minimum description length) 13, 14] which is known to be superior to Akaike s AIC in some respects. Naturally, there have been attempts to apply MDL to computer vision and robotics prob lems [3, 9, 12, 16]. The purpose of this paper is to point out that some of these attempts are based on a misunderstanding of the fundamental structure of geometric problems. The key fact is the duality that the asymptotic properties of statistical inference in the limit of an infinite number of observations hold ....
B. A. Maxwell, Segmentation and interpretation of multicolored objects with highlights, Cornput. Vision Image Understand., 77-1 (2000), 1 24.
Calibration of a Moving Camera Using a Planar Pattern.. - Matsunaga, Kanatani (1999) (2 citations) (Correct)
....or model, and adopt the best one. In this paper, we use the geometric AIC introduced by Kanatani [12, 14] and the geometric MDL to be defined shortly as the model selection criterion. The geometric MDL we use is different from the traditional MDL used in statistics and some vision applications [8, 11, 21, 22, 31]. We compare the per formances of the geometric AIC and the geometric MDL by doing numerical simulations mid real image experiments. 2 Basic Principle We fix a XYZ world coordinate system in the scene and place a planar pattern in paxallel to the XY plane at a known distance d. We imagine a ....
....final form is as follows (we omit the details of the code length analysis) G MDL = J ke 2 log e 2. 14) We call this criterion the geometric MDL 4. This form can also be obtained from Pdssanen s MDL by replacing by 1 e 2 and is different from any MDLs used in statistics and vision applications [8, 11, 21, 22, 31] in that ours does not contain the logarithm of the number of the data. 6.2 Degeneracy detection If degeneracy occurs, the confidence interval (12) expands infinitely wide if no noise exist. In the presence of noise, it has a finite width. We decide that degen eracy has occurred if the ....
B. A. Maxwell, Segmentation and interpretation of multicolored objects with highlights, Cornput. Vision Image Understand., 77-1 (2000), I 24.
Towards Object-Based Retrieval for Image Libraries - Bruce Maxwell Swarthmore (2001) Self-citation (Maxwell) (Correct)
No context found.
B. A. Maxwell and S. A. Shafer, "Segmentation and Interpretation of Multi-colored Objects with Highlights", Computer Vision and Image Understanding, Vol.77, No.1 January 2000, pp. 1-24.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC