Comparison of stochastic and deterministic solution methods in Bayesian estimation of 2D motion
Image and Vision ComputingPublished 1 November 1990
Janusz Konrad, Éric Dubois
Citations25
SJR quartileQ1
SJR score0.79
SNIP1.33
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
The estimation of 2D motion from spatio-temporally sampled image sequences is discussed, concentrating on the optimization aspect of the problem formulated through a Bayesian framework based on Markov random field (MRF) models. First, the Maximum A Posteriori Probability (MAP) formulation for motion estimation over discrete and continuous state spaces is reviewed along with the solution method using simulated annealing (SA). Then, instantaneous 'freezing' is applied to the stochastic algorithms resulting in well known deterministic methods. The stochastic algorithms are compared with their deterministic approximations over image sequences with natural data and synthetic as well as natural motion.
Keywords
Computer Science
ScienceOptimization by Simulated Annealing
44,600 Citations1983Scott Kirkpatrick, C. D. Gelatt +1 more
A detailed analogy with annealing in solids provides a framework for optimization of the properties of very large and complex systems.
IEEE Transactions on Pattern Analysis and Machine IntelligenceStochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
17,980 Citations1984Stuart Geman, Donald Geman
The analogy between images and statistical mechanics systems is made and the analogous operation under the posterior distribution yields the maximum a posteriori (MAP) estimate of the image given the degraded observations, creating a highly parallel ``relaxation'' algorithm for MAP estimation.
Artificial IntelligenceDetermining optical flow
9,940 Citations1981Berthold K. P. Horn, Brian G. Schunck
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Determining Optical Flow</title>
7,496 Citations1981Berthold K. P. Horn, Brian G. Schunck
An iterative implementation is shown which successfully computes the optical flow for a number of synthetic image sequences and is robust in that it can handle image sequences that are quantized rather coarsely in space and time.
Journal of the Royal Statistical Society Series B (Statistical Methodology)On the Statistical Analysis of Dirty Pictures
4,228 Citations1986Julian Besag
A simple, iterative method of reconstruction is proposed, which does not depend on these large-scale characteristics of the random field, and is illustrated by computer simulations in which the original scene is not directly related to the assumed random field.
IEEE Transactions on Acoustics Speech and Signal ProcessingCubic convolution interpolation for digital image processing
3,615 Citations1981Robert G. Keys
It can be shown that the order of accuracy of the cubic convolution method is between that of linear interpolation and that of cubic splines.
IEEE Transactions on Pattern Analysis and Machine IntelligenceAn Investigation of Smoothness Constraints for the Estimation of Displacement Vector Fields from Image Sequences
766 Citations1986Hans-Hellmut Nagel, W. Enkelmann
Results are reported of an investigation about how an ``oriented smoothness'' constraint may be formulated and evaluated about how variations of the displacement vector field only in directions with small or no variation of gray values are restricted.
Proceedings of the IEEEImage data compression: A review
711 Citations1981Anil K. Jain
A large variety of algorithms for image data compression are considered, starting with simple techniques of sampling and pulse code modulation (PCM) and state of the art algorithms for two-dimensional data transmission are reviewed.
Bell System Technical JournalMotion-Compensated Television Coding: Part I
463 Citations1979Arun N. Netravali, J. D. Robbins
Methods of estimating displacements of moving objects from one frame to the next in a television scene and using such displacements for frame-to-frame prediction by a recursive algorithm are presented which make it attractive for hardware implementation.
SIAM Journal on Control and OptimizationDiffusions for Global Optimization
306 Citations1986Stuart Geman, Chii-Ruey Hwang
IEEE Transactions on Pattern Analysis and Machine IntelligenceScene Segmentation from Visual Motion Using Global Optimization
300 Citations1987David W. Murray, Bernard F. Buxton
To compute the flow predicted by the segmentation, a recent method for reconstructing the motion and orientation of planar surface facets is used and the search for the globally optimal segmentation is performed using simulated annealing.
Artificial IntelligenceComputations underlying the measurement of visual motion
300 Citations1984Ellen C. Hildreth
A theoretical analysis of the conditions under which this computation of an instantaneous two-dimensional velocity field yields the correct velocity field suggests that the solution is physically plausible; empirical studies show the results to be consistent with human motion perception.
ComputerComputing motion using analog and binary resistive networks
295 Citations1988James Hutchinson, Christof Koch +2 more
The authors describe recent developments in the theory of early vision that led from the formulation of the motion problem as an ill-posed one to its solution by minimizing certain 'cost' functions by mapping onto simple analog and digital resistive networks.
International Journal of Computer VisionThe theory and practice of Bayesian image labeling
271 Citations1990Paul B. Chou, Christopher M. Brown
A new inference method, Highest Confidence First (HCF) estimation, is used to infer a unique labeling from the a posteriori distribution that is consistent with both prior knowledge and evidence.
Proceedings of the IEEEThe sampling and reconstruction of time-varying imagery with application in video systems
270 Citations1985Éric Dubois
An overview of the theory of sampling and reconstruction of multidimensional signals, including the role of the camera and display apertures, and the human visual system is presented and a class of nonlinear interpolation algorithms which adapt to the motion in the scene is presented.
International Journal of Computer VisionStochastic stereo matching over scale
238 Citations1989Stephen T. Barnard
A stochastic optimization approach to stereo matching is presented, which provides a dense array of disparities, eliminating the need for interpolation.
Lecture notes in computer scienceParallel and deterministic algorithms from MRFs: Surface reconstruction and integration
86 Citations1990Davi Geiger, Federico Girosi
Deterministic approximations to MRFs models are derived and one of the considered models for image reconstruction is shown to give in a natural way the graduate non-convexity algorithm proposed by Blake and Zisserman.
Computer Vision Graphics and Image ProcessingA spatio-temporal gradient method for estimating the displacement field in time-varying imagery
50 Citations1983Réal Paquin, Éric Dubois
Estimation of image motion fields: Bayesian formulation and stochastic solution
35 Citations2003Janusz Konrad, Éric Dubois
It is shown that motion estimation, an ill-posed problem, can be regularized by means of a Bayesian estimation approach and a probabilistic formulation for motion estimation in images and a stochastic algorithm for minimization of the associated objective function.
International Conference on Acoustics, Speech, and Signal ProcessingMotion detection in an image sequence using Gibbs distributions
18 Citations2003Patrick Bouthémy, Patrick Lalande
The authors address the problem of motion detection in an image sequence from the variations in time of the intensity distribution by the joint treatment of the detection of temporal changes and the reconstruction of mobile object masks according to a probabilistic formulation.
International Conference on Acoustics, Speech, and Signal ProcessingMotion estimation and segmentation using a global Bayesian approach
16 Citations2002Fabrice Heitz, Patrick Bouthémy
It is shown that optic flow estimation and segmentation can be expressed, within a Bayesian decision framework, as a global estimation problem, involving complex 3D motions and occlusions.
Computational Aspects Of Determining Optical Flow
8 Citations2005D. Lee, Anargyros Papageorgiou +1 more
It is shown that both free and Neuman boundary problems are ill-conditioned, and are not appropriate for optical flow computation, and the Chebyshev method is proposed for the computation.
OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)Parallel and deterministic algorithms from MRFs (Markov Random Fields): Surface reconstruction and integration. Memorandum report
6 Citations1989Davi Geiger, Federico Girosi
eScholarship@McGill (McGill)Bayesian estimation of motion fields from image sequences
4 Citations1989Konrad, Janusz