The modeling of spatial discontinuities for problems such as surface recovery, segmentation, image reconstruction, and optical flow has been intensely studied in computer vision. While "line-process " models of discontinuitieshave received a great deal of attention, there has been recent interest in the use of robust statistical techniques to account for discontinuities. This paper unifies the two approaches. To achieve this we generalize the notion of a "line process" to that of an analog "outlier process " and show how a problem formulated in terms of outlier processes can be viewed in terms of robust statistics. We also characterize a class of robust statistical problems for which an equivalent outlier-process formulation exists and give a straightforward method for converting a robust estimation problem into an outlier-process formulation. We show how prior assumptions about the spatial structure of outliers can be expressed as constraints on the recovered analog outlier processes and how traditional continuation methods can be extended to the explicit outlier-process formulation. These results indicate that the outlierprocesses approach provides a general framework which subsumes the traditional line-process approaches as well as a wide class of robust estimation problems. Examples in surface reconstruction, image segmentation, and optical flow are presented to illustrate the use of outlier processes and to show how the relationship between outlier processes and robust statistics can be exploited. An appendix provides a catalog of common robust error norms and their equivalent outlier-process formulations.
|
2405
|
Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images
– Geman, Geman
- 1984
|
|
1728
|
A Computation Approach to Edge Detection
– Canny
- 1986
|
|
1076
|
Determining optical flow
– Horn, Schunck
- 1981
|
|
940
|
Robust Statistics
– Huber
- 1981
|
|
706
|
Robust Regression and Outlier Detection
– Rousseeuw, Leroy
- 1987
|
|
683
|
Scale-space and edge detection using anisotropic diffusion
– Perona, Malik
- 1990
|
|
549
|
Visual reconstruction
– Blake, Zisserman
- 1987
|
|
326
|
Robust Statistics: The Approach Based on Influence Functions
– Hampel, Ronchetti, et al.
- 1986
|
|
214
|
Regularization of inverse visual problems involving discontinuities
– Terzopoulos
- 1986
|
|
211
|
Robust regression methods for computer vision: A review
– Meer, Mintz, et al.
- 1991
|
|
188
|
Constructing simple stable description for image partitioning
– Leclerc
- 1989
|
|
185
|
Constrained restoration and the recovery of discontinuities
– Geman, Reynolds
- 1992
|
|
166
|
Modelling and segmentation of noisy and textured images using Gibbs random field
– Derin, Elliot
- 1987
|
|
162
|
Boundary detection by minimizing functionals
– Mumford, Shah
- 1985
|
|
152
|
Probabilistic solution of ill-posed problems in computational vision
– Marroquin, Mitter, et al.
- 1987
|
|
144
|
Ill-posed problems in early vision
– Bertero, Poggio, et al.
- 1988
|
|
129
|
Parallel and deterministic algorithms from MRF’s: Surface reconstruction
– Geiger, Girosi
- 1991
|
|
125
|
Scene segmentation from visual motion using global optimization
– Murray, Buxton
- 1987
|
|
121
|
Statistical methods for tomographic image reconstruction
– Gemen, McClure
- 1987
|
|
115
|
Boundary detection by constrained optimization
– Geman, Geman, et al.
- 1990
|
|
107
|
The robust estimation of optical flow
– Black, Anandan
- 1993
|
|
88
|
C.M.Brown, “The theory and practice of bayesian image labeling
– Chou
|
|
78
|
Robust dynamic motion estimation over time
– Black, Anandan
- 1991
|
|
73
|
A common framework for image segmentation
– Geiger, Yuille
- 1991
|
|
68
|
Robust Incremental Optical Flow
– Black
- 1992
|
|
65
|
Lectures on Cauchy’s problems in linear partial differential equations (Yale
– Hadamard
- 1923
|
|
59
|
Linear feature extraction and description
– Nevatia, Babu
- 1980
|
|
54
|
A generalized EM algorithm for 3-d Bayesian reconstruction from Poisson data using Gibbs priors
– Hebert, Leahy
- 1989
|
|
51
|
Image flow segmentation and estimation by constraint line clustering
– Schunck
- 1989
|
|
41
|
and L.T.Watson, \Robust Window Operators
– Besl
- 1988
|
|
34
|
A two-stage algorithm for discontinuity-preserving surface reconstruction
– Sinha, Schunck
- 1992
|
|
32
|
The fitting of power series, meaning polynomials, illustrated on band-spectroscopic data
– Beaton, Tukey
- 1974
|
|
32
|
Robust Procedures in Multivariate Analysis I : Robust Covariance Estimation
– Campbell
- 1980
|
|
29
|
Regularization of discontinuous flow fields
– Shulman, Hervé
- 1989
|
|
24
|
The robust estimation of multiple motions: Affine and piecewise-smooth flow fields
– Black, Anandan
- 1993
|
|
24
|
Reliability analysis of parameter estimation in linear models with applications to mensuration problems in computer vision
– Forstner
- 1987
|
|
24
|
Analog hardware for detecting discontinuities in early vision
– Harris, Koch, et al.
- 1990
|
|
19
|
Robust Estimates of Location: Survey and Advances
– Andrews, Bickel, et al.
- 1972
|
|
17
|
Relaxation and its Role in Vision
– Hinton
- 1978
|
|
17
|
Multigrid Bayesian estimation of image motion fields using stochastic relaxation
– Konrad, Dubois
- 1988
|
|
13
|
Analysis of different robust methods for pose refinement
– Kumar, Hanson
- 1990
|
|
13
|
Segmentation by nonlinear diffusion
– Shah
- 1991
|
|
11
|
A continuation method for image estimation using the adiabatic approximation
– Rangarajan, Chellappa
- 1993
|
|
10
|
A continuation method for emission tomography
– Lee, Rangarajan, et al.
- 1993
|
|
9
|
Robust computational vision
– Schunck
- 1990
|
|
7
|
Motion estimation and segmentation
– Tian, Shah
- 1996
|
|
6
|
Robust statistical methods for building classification procedures
– Chen, Schunck
- 1990
|
|
5
|
On robust edge detection
– Lui, Schunck, et al.
- 1990
|
|
5
|
Robust recovery of piecewise polynomial image structure
– Meer, Mintz, et al.
- 1990
|
|
5
|
Robust dynamic stereo for incremental disparity map refinement
– Tirumalai, Schunck, et al.
- 1990
|
|
