Abstract—This article addresses two important themes in early visual computation: First, it presents a novel theory for learning the universal statistics of natural images—a prior model for typical cluttered scenes of the world—from a set of natural images, and, second, it proposes a general framework of designing reaction-diffusion equations for image processing. We start by studying the statistics of natural images including the scale invariant properties, then generic prior models were learned to duplicate the observed statistics, based on the minimax entropy theory studied in two previous papers. The resulting Gibbs distributions have potentials of
|
2322
|
Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images
– Geman, Geman
- 1984
|
|
658
|
Scale-space and edge detection using anisotropic diffusion
– Perona, Malik
- 1990
|
|
546
|
A maximum entropy approach to natural language processing
– Berger, Pietra, et al.
- 1996
|
|
536
|
Visual Reconstruction
– Blake, Zisserman
- 1987
|
|
439
|
Theory of communication
– Gabor
- 1946
|
|
433
|
Optimal approximations by piecewise smooth functions and associated variational problems
– Mumford, Shah
- 1989
|
|
425
|
Solution of Ill-posed Problems
– Tikhonov, Arsenin
- 1977
|
|
358
|
Uncertainty relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters
– Daugman
- 1985
|
|
357
|
Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation
– Zhu, Yuille
- 1996
|
|
348
|
Relations between the statistics of natural images and the response properties of cortical cells
– Field
- 1987
|
|
209
|
Robust Regression Methods for Computer Vision
– Meer, Mintz, et al.
- 1991
|
|
159
|
Generating textures on arbitrary surfaces using reactiondiffusion
– Turk
- 1991
|
|
146
|
Image analysis, random fields, and dynamic Monte Carlo methods : a mathematical introduction. Berlin
– Winkler
- 1995
|
|
140
|
Robust anisotropic diffusion
– Black, Sapiro, et al.
- 1998
|
|
134
|
Minimax entropy principle and its application to texture modeling
– Zhu, Wu, et al.
- 1997
|
|
130
|
On the unification of line processes, outlier rejection, and robust statistics with applications in early vision
– Black, Rangarajan
- 1996
|
|
96
|
Reaction-diffusion textures
– Kass
- 1991
|
|
72
|
A common framework for image segmentation
– Geiger, Yuille
- 1991
|
|
58
|
Nonlinear image filtering with edge and corner enhancement
– Nitzberg, Shiota
- 1992
|
|
56
|
Multilevel computational processes for visual surface reconstruction
– Terzopoulos
- 1983
|
|
50
|
A renormalization group approach to image processing problems
– Gidas
- 1989
|
|
48
|
A Theory of Networks for Approximation and
– Poggio, Girosi
- 1989
|
|
33
|
Diffusions for global optimization
– Geman, Hwang
- 1986
|
|
29
|
Parallel and deterministic algorithms for mrfs: surface reconstruction
– Geiger, Girosi
- 1991
|
|
29
|
A model for the formation of ocular dominance stripes
– Swindale
- 1980
|
|
23
|
Statistics of Natural images: Scaling in the
– Ruderman, Bialek
- 1994
|
|
15
|
Learning Generic Prior Models for Visual Computation
– Zhu, Mumford
- 1997
|
|
9
|
M-lattice: from morphogenesis to image processing
– Sherstinsky, Picard
- 1996
|
|
8
|
The Theory and Applications of Reaction-Diffusion Equations
– Grindrod
- 1996
|
|
8
|
Probabilistic Solution of Ill-Posed Problems in
– Marroquin, Mitter, et al.
- 1987
|
|
8
|
The chemical basis of morphogenesis,” Philosophical
– Turing
- 1952
|
|
5
|
A pre-pattern formation mechanism for mammalian coat markings
– Murray
- 1981
|
|
5
|
A general framework for geometry-driven evolution equations
– Niessen
- 1997
|
|
4
|
Constrained restoration and the recover of discontinuities
– Geman, Reynoids
- 1992
|
|
3
|
Efficiency of Model Human Image
– Watson
- 1987
|
|
1
|
On Sampling Methods and Annealing Algorithms,” Markov Random Fields—Theory and Applications
– Gelfand, Mitter
- 1993
|
|
1
|
Deformations I: The Components of Two-Dimensional Shape and
– Kimia, Tannebaum, et al.
- 1997
|
|
1
|
The Renormalization Group: Critical Phenomena and the Knodo Problem
– Wilson
- 1975
|
|
1
|
Random Fields, and Minimax Entropy (FRAME): Towards a Unified Theory for Texture Modeling
– Zhu, Wu, et al.
- 1996
|
