Results 1  10
of
145
Interactive Graph Cuts for Optimal Boundary & Region Segmentation of Objects in ND Images
, 2001
"... In this paper we describe a new technique for general purpose interactive segmentation of Ndimensional images. The user marks certain pixels as “object” or “background” to provide hard constraints for segmentation. Additional soft constraints incorporate both boundary and region information. Graph ..."
Abstract

Cited by 1010 (20 self)
 Add to MetaCart
In this paper we describe a new technique for general purpose interactive segmentation of Ndimensional images. The user marks certain pixels as “object” or “background” to provide hard constraints for segmentation. Additional soft constraints incorporate both boundary and region information. Graph cuts are used to find the globally optimal segmentation of the Ndimensional image. The obtained solution gives the best balance of boundary and region properties among all segmentations satisfying the constraints. The topology of our segmentation is unrestricted and both “object” and “background” segments may consist of several isolated parts. Some experimental results are presented in the context of photo/video editing and medical image segmentation. We also demonstrate an interesting Gestalt example. A fast implementation of our segmentation method is possible via a new maxflow algorithm in [2].
Random walks for image segmentation
 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
, 2006
"... A novel method is proposed for performing multilabel, interactive image segmentation. Given a small number of pixels with userdefined (or predefined) labels, one can analytically and quickly determine the probability that a random walker starting at each unlabeled pixel will first reach one of the ..."
Abstract

Cited by 387 (21 self)
 Add to MetaCart
(Show Context)
A novel method is proposed for performing multilabel, interactive image segmentation. Given a small number of pixels with userdefined (or predefined) labels, one can analytically and quickly determine the probability that a random walker starting at each unlabeled pixel will first reach one of the prelabeled pixels. By assigning each pixel to the label for which the greatest probability is calculated, a highquality image segmentation may be obtained. Theoretical properties of this algorithm are developed along with the corresponding connections to discrete potential theory and electrical circuits. This algorithm is formulated in discrete space (i.e., on a graph) using combinatorial analogues of standard operators and principles from continuous potential theory, allowing it to be applied in arbitrary dimension on arbitrary graphs.
Graph Cuts and Efficient ND Image Segmentation
, 2006
"... Combinatorial graph cut algorithms have been successfully applied to a wide range of problems in vision and graphics. This paper focusses on possibly the simplest application of graphcuts: segmentation of objects in image data. Despite its simplicity, this application epitomizes the best features ..."
Abstract

Cited by 307 (7 self)
 Add to MetaCart
Combinatorial graph cut algorithms have been successfully applied to a wide range of problems in vision and graphics. This paper focusses on possibly the simplest application of graphcuts: segmentation of objects in image data. Despite its simplicity, this application epitomizes the best features of combinatorial graph cuts methods in vision: global optima, practical efficiency, numerical robustness, ability to fuse a wide range of visual cues and constraints, unrestricted topological properties of segments, and applicability to ND problems. Graph cuts based approaches to object extraction have also been shown to have interesting connections with earlier segmentation methods such as snakes, geodesic active contours, and levelsets. The segmentation energies optimized by graph cuts combine boundary regularization with regionbased properties in the same fashion as MumfordShah style functionals. We present motivation and detailed technical description of the basic combinatorial optimization framework for image segmentation via s/t graph cuts. After the general concept of using binary graph cut algorithms for object segmentation was first proposed and tested in Boykov and Jolly (2001), this idea was widely studied in computer vision and graphics communities. We provide links to a large number of known extensions based on iterative parameter reestimation and learning, multiscale or hierarchical approaches, narrow bands, and other techniques for demanding photo, video, and medical applications.
Combining topdown and bottomup segmentation
 In Proceedings IEEE workshop on Perceptual Organization in Computer Vision, CVPR
, 2004
"... In this work we show how to combine bottomup and topdown approaches into a single figureground segmentation process. This process provides accurate delineation of object boundaries that cannot be achieved by either the topdown or bottomup approach alone. The topdown approach uses object represen ..."
Abstract

Cited by 191 (2 self)
 Add to MetaCart
(Show Context)
In this work we show how to combine bottomup and topdown approaches into a single figureground segmentation process. This process provides accurate delineation of object boundaries that cannot be achieved by either the topdown or bottomup approach alone. The topdown approach uses object representation learned from examples to detect an object in a given input image and provide an approximation to its figureground segmentation. The bottomup approach uses imagebased criteria to define coherent groups of pixels that are likely to belong together to either the figure or the background part. The combination provides a final segmentation that draws on the relative merits of both approaches: The result is as close as possible to the topdown approximation, but is also constrained by the bottomup process to be consistent with significant image discontinuities. We construct a global cost function that represents these topdown and bottomup requirements. We then show how the global minimum of this function can be efficiently found by applying the sumproduct algorithm. This algorithm also provides a confidence map that can be used to identify image regions where additional topdown or bottomup information may further improve the segmentation. Our experiments show that the results derived from the algorithm are superior to results given by a pure topdown or pure bottomup approach. The scheme has broad applicability, enabling the combined use of a range of existing bottomup and topdown segmentations. 1.
An ultrafast usersteered image segmentation paradigm: livewireonthefly
 IEEE Transactions on Medical Imaging
, 2000
"... Abstract—We have been developing general user steered image segmentation strategies for routine use in applications involving a large number of data sets. In the past, we have presented three segmentation paradigms: live wire, live lane, and a threedimensional (3D) extension of the livewire meth ..."
Abstract

Cited by 141 (16 self)
 Add to MetaCart
Abstract—We have been developing general user steered image segmentation strategies for routine use in applications involving a large number of data sets. In the past, we have presented three segmentation paradigms: live wire, live lane, and a threedimensional (3D) extension of the livewire method. In this paper, we introduce an ultrafast livewire method, referred to as live wire on the fly, for further reducing user’s time compared to the basic livewire method. In live wire, 3D/fourdimensional (4D) object boundaries are segmented in a slicebyslice fashion. To segment a twodimensional (2D) boundary, the user initially picks a point on the boundary and all possible minimumcost paths from this point to all other points in the image are computed via Dijkstra’s algorithm. Subsequently, a live wire is displayed in real time from the initial point to any subsequent position taken by the cursor. If the cursor is close to the desired boundary, the live wire snaps on to the boundary. The cursor is then deposited and a new livewire segment is found next. The entire 2D boundary is specified via a
Alpha Estimation in Natural Images
, 2000
"... Many boundaries between objects in the world project onto curves in an image. However, boundaries involving natural objects (e.g., trees, hair, water, smoke) are often unworkable under this model because many pixels receive light from more than one object. We propose a technique for estimating alpha ..."
Abstract

Cited by 118 (0 self)
 Add to MetaCart
Many boundaries between objects in the world project onto curves in an image. However, boundaries involving natural objects (e.g., trees, hair, water, smoke) are often unworkable under this model because many pixels receive light from more than one object. We propose a technique for estimating alpha, the proportion in which two colors mix to produce a color at the boundary. The technique extends blue screen matting to backgrounds that have almost arbitrary color distributions, though coarse knowledge of the boundary's location is required. Results show a number of different objects moved from one image to another while maintaining naturalism. 1. Introduction The popularity of imagebased rendering techniques has led to increased interest in extracting objects from one image to be placed in another. When boundaries are in focus and are well modeled by a set of edges meeting at corners, a reasonable effect can be obtained by cutting and pasting followed by a smoothing operation along t...
A Seeded Image Segmentation Framework Unifying Graph Cuts And Random Walker Which Yields A New Algorithm
 ICCV
, 2007
"... In this work, we present a common framework for seeded image segmentation algorithms that yields two of the leading methods as special cases The Graph Cuts and the Random Walker algorithms. The formulation of this common framework naturally suggests a new, third, algorithm that we develop here. Spe ..."
Abstract

Cited by 97 (9 self)
 Add to MetaCart
In this work, we present a common framework for seeded image segmentation algorithms that yields two of the leading methods as special cases The Graph Cuts and the Random Walker algorithms. The formulation of this common framework naturally suggests a new, third, algorithm that we develop here. Specifically, the former algorithms may be shown to minimize a certain energy with respect to either an ℓ1 or an ℓ2 norm. Here, we explore the segmentation algorithm defined by an ℓ ∞ norm, provide a method for the optimization and show that the resulting algorithm produces an accurate segmentation that demonstrates greater stability with respect to the number of seeds employed than either the Graph Cuts or Random Walker methods.
Interactive organ segmentation using graph cuts
 In Medical Image Computing and ComputerAssisted Intervention
, 2000
"... Abstract. An Ndimensional image is divided into “object ” and “background” segments using a graph cut approach. A graph is formed by connecting all pairs of neighboring image pixels (voxels) by weighted edges. Certain pixels (voxels) have to be a priori identified as object or background seeds prov ..."
Abstract

Cited by 80 (1 self)
 Add to MetaCart
(Show Context)
Abstract. An Ndimensional image is divided into “object ” and “background” segments using a graph cut approach. A graph is formed by connecting all pairs of neighboring image pixels (voxels) by weighted edges. Certain pixels (voxels) have to be a priori identified as object or background seeds providing necessary clues about the image content. Our objective is to find the cheapest way to cut the edges in the graph so that the object seeds are completely separated from the background seeds. If the edge cost is a decreasing function of the local intensity gradient then the minimum cost cut should produce an object/background segmentation with compact boundaries along the high intensity gradient values in the image. An efficient, globally optimal solution is possible via standard mincut/maxflow algorithms for graphs with two terminals. We applied this technique to interactively segment organs in various 2D and 3D medical images. 1
Mesh scissoring with minima rule and part salience
, 2005
"... This paper presents an intelligent scissoring operator for meshes. Unlike common approaches that segment a mesh using clustering schemes, here we introduce a method that concentrates on the contours for cutting. Our approach is based on the minima rule and part salience theory from the cognitive the ..."
Abstract

Cited by 55 (2 self)
 Add to MetaCart
This paper presents an intelligent scissoring operator for meshes. Unlike common approaches that segment a mesh using clustering schemes, here we introduce a method that concentrates on the contours for cutting. Our approach is based on the minima rule and part salience theory from the cognitive theory. The minima rule states that human perception usually divides a surface into parts along the concave discontinuity of the tangent plane. The part salience theory provides factors which determine the salience of segments. Our method first extracts features to find candidate contours based on the minima rule. Subsequently, these open contours are prioritized to select the most salient one. Then, the selected open contour is automatically completed to form a loop around a specific part of the mesh. This loop is used as the initial position of a 3D geometric snake. Before applying a snake, we measure the part salience of the segments obtained by the completed contour. If conditions for the salience are not met, the contour is rejected. Otherwise, the snake moves by relaxation until it settles to define the final scissoring position. In this paper, we focus on a fully automatic scissoring scheme; nevertheless, we also report on semiautomatic user interfaces for intelligent scissoring which are easy to use and intuitive.