Results 1  10
of
16
Fast approximate energy minimization with label costs
, 2010
"... The αexpansion algorithm [7] has had a significant impact in computer vision due to its generality, effectiveness, and speed. Thus far it can only minimize energies that involve unary, pairwise, and specialized higherorder terms. Our main contribution is to extend αexpansion so that it can simult ..."
Abstract

Cited by 44 (6 self)
 Add to MetaCart
The αexpansion algorithm [7] has had a significant impact in computer vision due to its generality, effectiveness, and speed. Thus far it can only minimize energies that involve unary, pairwise, and specialized higherorder terms. Our main contribution is to extend αexpansion so that it can simultaneously optimize “label costs ” as well. An energy with label costs can penalize a solution based on the set of labels that appear in it. The simplest special case is to penalize the number of labels in the solution. Our energy is quite general, and we prove optimality bounds for our algorithm. A natural application of label costs is multimodel fitting, and we demonstrate several such applications in vision: homography detection, motion segmentation, and unsupervised image segmentation. Our C++/MATLAB implementation is publicly available.
Clustering under prior knowledge with application to image segmentation
 Advances in Neural Information Processing Systems 19
, 2007
"... This paper proposes a new approach to modelbased clustering under prior knowledge. The proposed formulation can be interpreted from two different angles: as penalized logistic regression, where the class labels are only indirectly observed (via the probability density of each class); as finite mixt ..."
Abstract

Cited by 9 (0 self)
 Add to MetaCart
This paper proposes a new approach to modelbased clustering under prior knowledge. The proposed formulation can be interpreted from two different angles: as penalized logistic regression, where the class labels are only indirectly observed (via the probability density of each class); as finite mixture learning under a grouping prior. To estimate the parameters of the proposed model, we derive a (generalized) EM algorithm with a closedform Estep, in contrast with other recent approaches to semisupervised probabilistic clustering which require Gibbs sampling or suboptimal shortcuts. We show that our approach is ideally suited for image segmentation: it avoids the combinatorial nature Markov random field priors, and opens the door to more sophisticated spatial priors (e.g., waveletbased) in a simple and computationally efficient way. Finally, we extend our formulation to work in unsupervised, semisupervised, or discriminative modes. 1
Energybased Geometric MultiModel Fitting
, 2010
"... Geometric model fitting is a typical chicken&egg problem: data points should be clustered based on geometric proximity to models whose unknown parameters must be estimated at the same time. Most existing methods, including generalizations of RANSAC, greedily search for models with most inliers (wi ..."
Abstract

Cited by 9 (2 self)
 Add to MetaCart
Geometric model fitting is a typical chicken&egg problem: data points should be clustered based on geometric proximity to models whose unknown parameters must be estimated at the same time. Most existing methods, including generalizations of RANSAC, greedily search for models with most inliers (within a threshold) ignoring overall classification of points. We formulate geometric multimodel fitting as an optimal labeling problem with a global energy function balancing geometric errors and regularity of inlier clusters. Regularization based on spatial coherence (on some nearneighbor graph) and/or label costs is NP hard. Standard combinatorial algorithms with guaranteed approximation bounds (e.g. αexpansion) can minimize such regularization energies over a finite set of labels, but they are not directly applicable to a continuum of labels, e.g. R 2 in line fitting. Our proposed approach (PEARL) combines model sampling from data points as in RANSAC with iterative reestimation of inliers and models parameters based on a global regularization functional. This technique efficiently explores the continuum of labels in the context of energy minimization. In practice, PEARL converges to a good quality local minima of the energy automatically selecting a small number of models that best explain the whole data set. Our tests demonstrate that our energybased approach significantly improves the current state of the art in geometric model fitting currently dominated by various greedy generalizations of RANSAC.
Incorporating nonmotion cues into 3d motion segmentation
 in Proc. of ECCV
, 2006
"... We address the problem of segmenting an image sequence into rigidly moving 3D objects. An elegant solution to this problem is the multibody factorization approach in which the measurement matrix is factored into lower rank matrices. Despite progress in factorization algorithms, the performance is st ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
We address the problem of segmenting an image sequence into rigidly moving 3D objects. An elegant solution to this problem is the multibody factorization approach in which the measurement matrix is factored into lower rank matrices. Despite progress in factorization algorithms, the performance is still far from satisfactory and in scenes with missing data and noise, most existing algorithms fail. In this paper we propose a method for incorporating 2D nonmotion cues (such as spatial coherence) into multibody factorization. We formulate the problem in terms of constrained factor analysis and use the EM algorithm to find the segmentation. We show that adding these cues improves performance in real and synthetic sequences. 1
Parameterfree spatial data mining using MDL
 In 5th International Conference on Data Mining (ICDM
, 2005
"... Consider spatial data consisting of a set of binary features taking values over a collection of spatial extents (grid cells). We propose a method that simultaneously finds spatial correlation and feature cooccurrence patterns, without any parameters. In particular, we employ the Minimum Description ..."
Abstract

Cited by 7 (1 self)
 Add to MetaCart
Consider spatial data consisting of a set of binary features taking values over a collection of spatial extents (grid cells). We propose a method that simultaneously finds spatial correlation and feature cooccurrence patterns, without any parameters. In particular, we employ the Minimum Description Length (MDL) principle coupled with a natural way of compressing regions. This defines what “good” means: a feature cooccurrence pattern is good, if it helps us better compress the set of locations for these features. Conversely, a spatial correlation is good, if it helps us better compress the set of features in the corresponding region. Our approach is scalable for large datasets (both number of locations and of features). We evaluate our method on both real and synthetic datasets. 1
Bayesian image segmentation using waveletbased priors
 Proc. IEEE Conf. Computer Vision and Pattern Recognition  CVPR’2005
, 2005
"... This paper introduces a formulation which allows using waveletbased priors for image segmentation. This formulation can be used in supervised, unsupervised, or semisupervised modes, and with any probabilistic observation model (intensity, multispectral, texture). Our main goal is to exploit the wel ..."
Abstract

Cited by 6 (2 self)
 Add to MetaCart
This paper introduces a formulation which allows using waveletbased priors for image segmentation. This formulation can be used in supervised, unsupervised, or semisupervised modes, and with any probabilistic observation model (intensity, multispectral, texture). Our main goal is to exploit the wellknown ability of waveletbased priors to model piecewise smoothness (which underlies stateoftheart methods for denoising, coding, and restoration) and the availability of fast algorithms for waveletbased processing. The main obstacle to using waveletbased priors for segmentation is that they’re aimed at representing real values, rather than discrete labels, as needed for segmentation. This difficulty is sidestepped by the introduction of realvalued hidden fields, to which the labels are probabilistically related. These hidden fields, being unconstrained and realvalued, can be given any type of spatial prior, such as one based on wavelets. Under this model, Bayesian MAP segmentation is carried out by a (generalized) EM algorithm. Experiments on synthetic and real data testify for the adequacy of the approach. 1.
Spectral Rounding & Image Segmentation
, 2006
"... The task of assigning labels to pixels is central to computer vision. In automatic segmentation an algorithm assigns a label to each pixel where labels connote a shared property across pixels (e.g. color, bounding contour, texture). Recent approaches to image segmentation have formulated this label ..."
Abstract

Cited by 4 (1 self)
 Add to MetaCart
The task of assigning labels to pixels is central to computer vision. In automatic segmentation an algorithm assigns a label to each pixel where labels connote a shared property across pixels (e.g. color, bounding contour, texture). Recent approaches to image segmentation have formulated this labeling task as partitioning a graph derived from the image. We use spectral segmentation to denote the family of algorithms that seek a partitioning by processing the eigenstructure associated with image graphs. In this thesis we analyze current spectral segmentation algorithms and explain their performance, both practically and theoretically, on the Normalized Cuts (NCut) criterion. Further, we introduce a novel family of spectral graph partitioning methods, spectral rounding, and apply them to image segmentation tasks. Edge separators of a graph are produced by iteratively reweighting the edges until the graph disconnects into the prescribed number of components. At each iteration a small number of eigenvectors with small eigenvalue are computed and used to determine the reweighting. In this way spectral rounding directly produces discrete solutions where as current spectral algorithms must map the continuous eigenvectors to discrete
Bayesian image segmentation using Gaussian field priors
 In CVPR Workshop on Energy MinimizationMethods inComputer VisionandPatternRecognition
, 2005
"... Abstract. The goal of segmentation is to partition an image into a finite set of regions, homogeneous in some (e.g., statistical) sense, thus being an intrinsically discrete problem. Bayesian approaches to segmentation use priors to impose spatial coherence; the discrete nature of segmentation deman ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
Abstract. The goal of segmentation is to partition an image into a finite set of regions, homogeneous in some (e.g., statistical) sense, thus being an intrinsically discrete problem. Bayesian approaches to segmentation use priors to impose spatial coherence; the discrete nature of segmentation demands priors defined on discretevalued fields, thus leading to difficult combinatorial problems. This paper presents a formulation which allows using continuous priors, namely Gaussian fields, for image segmentation. Our approach completely avoids the combinatorial nature of standard Bayesian approaches to segmentation. Moreover, it’s completely general, i.e., itcanbeused in supervised, unsupervised, or semisupervised modes, with any probabilistic observation model (intensity, multispectral, or texture features). To use continuous priors for image segmentation, we adopt a formulation which is common in Bayesian machine learning: introduction of hidden fields to which the region labels are probabilistically related. Since these hidden fields are realvalued, we can adopt any type of spatial prior for continuousvalued fields, such as Gaussian priors. We show how, under this model, Bayesian MAP segmentation is carried out by a (generalized) EM algorithm. Experiments on synthetic and real data shows that the proposed approach performs very well at a low computational cost. 1
Interactive Segmentation with SuperLabels
 In Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR
, 2011
"... *authors contributed equally ..."