Results 1  10
of
42
Scale Mixtures of Gaussians and the Statistics of Natural Images
 in Adv. Neural Information Processing Systems
, 2000
"... The statistics of photographic images, when represented using multiscale (wavelet) bases, exhibit two striking types of nonGaussian behavior. First, the marginal densities of the coefficients have extended heavy tails. Second, the joint densities exhibit variance dependencies not captured by secon ..."
Abstract

Cited by 169 (17 self)
 Add to MetaCart
(Show Context)
The statistics of photographic images, when represented using multiscale (wavelet) bases, exhibit two striking types of nonGaussian behavior. First, the marginal densities of the coefficients have extended heavy tails. Second, the joint densities exhibit variance dependencies not captured by secondorder models. We examine properties of the class of Gaussian scale mixtures, and show that these densities can accurately characterize both the marginal and joint distributions of natural image wavelet coefficients. This class of model suggests a Markov structure, in which wavelet coefficients are linked by hidden scaling variables corresponding to local image structure. We derive an estimator for these hidden variables, and show that a nonlinear ``normalization'' procedure can be used to Gaussianize the coefficients.
Multiresolution markov models for signal and image processing
 Proceedings of the IEEE
, 2002
"... This paper reviews a significant component of the rich field of statistical multiresolution (MR) modeling and processing. These MR methods have found application and permeated the literature of a widely scattered set of disciplines, and one of our principal objectives is to present a single, coheren ..."
Abstract

Cited by 154 (19 self)
 Add to MetaCart
(Show Context)
This paper reviews a significant component of the rich field of statistical multiresolution (MR) modeling and processing. These MR methods have found application and permeated the literature of a widely scattered set of disciplines, and one of our principal objectives is to present a single, coherent picture of this framework. A second goal is to describe how this topic fits into the even larger field of MR methods and concepts–in particular making ties to topics such as wavelets and multigrid methods. A third is to provide several alternate viewpoints for this body of work, as the methods and concepts we describe intersect with a number of other fields. The principle focus of our presentation is the class of MR Markov processes defined on pyramidally organized trees. The attractiveness of these models stems from both the very efficient algorithms they admit and their expressive power and broad applicability. We show how a variety of methods and models relate to this framework including models for selfsimilar and 1/f processes. We also illustrate how these methods have been used in practice. We discuss the construction of MR models on trees and show how questions that arise in this context make contact with wavelets, state space modeling of time series, system and parameter identification, and hidden
Dynamic Model of Visual Recognition Predicts Neural Response Properties in the Visual Cortex
 Neural Computation
, 1995
"... this paper, we describe a hierarchical network model of visual recognition that explains these experimental observations by using a form of the extended Kalman filter as given by the Minimum Description Length (MDL) principle. The model dynamically combines inputdriven bottomup signals with expec ..."
Abstract

Cited by 114 (21 self)
 Add to MetaCart
this paper, we describe a hierarchical network model of visual recognition that explains these experimental observations by using a form of the extended Kalman filter as given by the Minimum Description Length (MDL) principle. The model dynamically combines inputdriven bottomup signals with expectationdriven topdown signals to predict current recognition state. Synaptic weights in the model are adapted in a Hebbian manner according to a learning rule also derived from the MDL principle. The resulting prediction/learning scheme can be viewed as implementing a form of the ExpectationMaximization (EM) algorithm. The architecture of the model posits an active computational role for the reciprocal connections between adjoining visual cortical areas in determining neural response properties. In particular, the model demonstrates the possible role of feedback from higher cortical areas in mediating neurophysiological effects due to stimuli from beyond the classical receptive field. Si
Efficient multiscale regularization with applications to the computation of optical flow
 IEEE Trans. Image Process
, 1994
"... AbsfruetA new approach to regularization methods for image processing is introduced and developed using as a vehicle the problem of computing dense optical flow fields in an image sequence. Standard formulations of this problem require the computationally intensive solution of an elliptic partial d ..."
Abstract

Cited by 106 (36 self)
 Add to MetaCart
(Show Context)
AbsfruetA new approach to regularization methods for image processing is introduced and developed using as a vehicle the problem of computing dense optical flow fields in an image sequence. Standard formulations of this problem require the computationally intensive solution of an elliptic partial differential equation that arises from the often used “smoothness constraint” ’yl”. regularization. The interpretation of the smoothness constraint is utilized as a “fractal prior ” to motivate regularization based on a recently introduced class of multiscale stochastic models. The solution of the new problem formulation is computed with an efficient multiscale algorithm. Experiments on several image sequences demonstrate the substantial computational savings that can be achieved due to the fact that the algorithm is noniterative and in fact has a per pixel computational complexity that is independent of image size. The new approach also has a number of other important advantages. Specifically, multiresolution flow field estimates are available, allowing great flexibility in dealing with the tradeoff between resolution and accuracy. Multiscale error covariance information is also available, which is of considerable use in assessing the accuracy of the estimates. In particular, these error statistics can be used as the basis for a rational procedure for determining the spatiallyvarying optimal reconstruction resolution. Furthermore, if there are compelling reasons to insist upon a standard smoothness constraint, our algorithm provides an excellent initialization for the iterative algorithms associated with the smoothness constraint problem formulation. Finally, the usefulness of our approach should extend to a wide variety of illposed inverse problems in which variational techniques seeking a “smooth ” solution are generally Used. I.
Random Cascades on Wavelet Trees and Their Use in Analyzing and Modeling Natural Images
 Applied and Computational Harmonic Analysis
, 2001
"... in signal and image processing, including image denoising, coding, and superresolution. # 2001 Academic Press 1. INTRODUCTION Stochastic models of natural images underlie a variety of applications in image processing and lowlevel computer vision, including image coding, denoising and 1 MW supp ..."
Abstract

Cited by 99 (15 self)
 Add to MetaCart
(Show Context)
in signal and image processing, including image denoising, coding, and superresolution. # 2001 Academic Press 1. INTRODUCTION Stochastic models of natural images underlie a variety of applications in image processing and lowlevel computer vision, including image coding, denoising and 1 MW supported by NSERC 1967 fellowship; AW and MW by AFOSR Grant F496209810349 and ONR Grant N0001491J1004. Address correspondence to MW. 2 ES supported by NSF Career Grant MIP9796040 and an Alfred P. Sloan fellowship. 89 10635203/01 $35.00 Copyright # 2001 by Academic Press All rights of reproduction in any form reserved. 90 WAINWRIGHT, SIMONCELLI, AND WILLSKY restoration, interpolation and synthesis. Accordingly, the past decade has witnessed an increasing amount of research devoted to developing stochastic models of images (e.g., [19, 38, 45, 48, 55]). Simultaneously, wavel
Image Processing with Multiscale Stochastic Models
, 1993
"... In this thesis, we develop image processing algorithms and applications for a particular class of multiscale stochastic models. First, we provide background on the model class, including a discussion of its relationship to wavelet transforms and the details of a twosweep algorithm for estimation. A ..."
Abstract

Cited by 34 (4 self)
 Add to MetaCart
(Show Context)
In this thesis, we develop image processing algorithms and applications for a particular class of multiscale stochastic models. First, we provide background on the model class, including a discussion of its relationship to wavelet transforms and the details of a twosweep algorithm for estimation. A multiscale model for the error process associated with this algorithm is derived. Next, we illustrate how the multiscale models can be used in the context of regularizing illposed inverse problems and demonstrate the substantial computational savings that such an approach offers. Several novel features of the approach are developed including a technique for choosing the optimal resolution at which to recover the object of interest. Next, we show that this class of models contains other widely used classes of statistical models including 1D Markov processes and 2D Markov random fields, and we propose a class of multiscale models for approximately representing Gaussian Markov random fields...
TreeBased Modeling and Estimation of Gaussian Processes on Graphs with Cycles
 In Neural Information Processing Systems 13
, 2000
"... We present the embedded trees algorithm, an iterative technique for estimation of Gaussian processes defined on arbitrary graphs. ..."
Abstract

Cited by 28 (12 self)
 Add to MetaCart
(Show Context)
We present the embedded trees algorithm, an iterative technique for estimation of Gaussian processes defined on arbitrary graphs.
Multiscale autoregressive models and wavelets
, 1999
"... The multiscale autoregressive (MAR) framework was introduced to support the development of optimal multiscale statistical signal processing. Its power resides in the fast and flexible algorithms to which it leads. While the MAR framework was originally motivated by wavelets, the link between these t ..."
Abstract

Cited by 26 (4 self)
 Add to MetaCart
The multiscale autoregressive (MAR) framework was introduced to support the development of optimal multiscale statistical signal processing. Its power resides in the fast and flexible algorithms to which it leads. While the MAR framework was originally motivated by wavelets, the link between these two worlds has been previously established only in the simple case of the Haar wavelet. The first contribution of this paper is to provide a unification of the MAR framework and all compactly supported wavelets as well as a new view of the multiscale stochastic realization problem. The second contribution of this paper is to develop waveletbased approximate internal MAR models for stochastic processes. This will be done by incorporating a powerful synthesis algorithm for the detail coefficients which complements the usual wavelet reconstruction algorithm for the scaling coefficients. Taking advantage of the statistical machinery provided by the MAR framework, we will illustrate the application of our models to samplepath generation and estimation from noisy, irregular, and sparse measurements.
Factor analysis using deltarule wakesleep learning
 Neural Computation
, 1997
"... We describe a linear network that models correlations between realvalued visible variables using one or more realvalued hidden variables — a factor analysis model. This model can be seen as a linear version of the “Helmholtz machine”, and its parameters can be learned using the “wakesleep ” metho ..."
Abstract

Cited by 26 (4 self)
 Add to MetaCart
We describe a linear network that models correlations between realvalued visible variables using one or more realvalued hidden variables — a factor analysis model. This model can be seen as a linear version of the “Helmholtz machine”, and its parameters can be learned using the “wakesleep ” method, in which learning of the primary “generative” model is assisted by a “recognition ” model, whose role is to fill in the values of hidden variables based on the values of visible variables. The generative and recognition models are jointly learned in “wake ” and “sleep ” phases, using just the delta rule. This learning procedure is comparable in simplicity to Oja’s version of Hebbian learning, which produces a somewhat different representation of correlations in terms of principal components. We argue that the simplicity of wakesleep learning makes factor analysis a plausible alternative to Hebbian learning as a model of activitydependent cortical plasticity. 1
Estimating Rates of Rare Events at Multiple Resolutions ABSTRACT
"... We consider the problem of estimating occurrence rates of rare events for extremely sparse data, using preexisting hierarchies to perform inference at multiple resolutions. In particular, we focus on the problem of estimating click rates for (webpage, advertisement) pairs (called impressions) where ..."
Abstract

Cited by 21 (6 self)
 Add to MetaCart
(Show Context)
We consider the problem of estimating occurrence rates of rare events for extremely sparse data, using preexisting hierarchies to perform inference at multiple resolutions. In particular, we focus on the problem of estimating click rates for (webpage, advertisement) pairs (called impressions) where both the pages and the ads are classified into hierarchies that capture broad contextual information at different levels of granularity. Typically the click rates are low and the coverage of the hierarchies is sparse. To overcome these difficulties we devise a sampling method whereby we analyze a specially chosen sample of pages in the training set, and then estimate click rates using a twostage model. The first stage imputes the number of (webpage, ad) pairs at all resolutions of the hierarchy to adjust for the sampling bias. The second stage estimates click rates at all resolutions after incorporating correlations among sibling nodes through a treestructured Markov model. Both models are scalable and suited to large scale data mining applications. On a realworld dataset consisting of 1/2 billion impressions, we demonstrate that even with 95 % negative (nonclicked) events in the training set, our method can effectively discriminate extremely rare events in terms of their click propensity.