## Efficient Graph-Based Image Segmentation

### Cached

### Download Links

Citations: | 527 - 1 self |

### BibTeX

@MISC{Felzenszwalb_efficientgraph-based,

author = {Pedro F. Felzenszwalb and Daniel P. Huttenlocher},

title = {Efficient Graph-Based Image Segmentation},

year = {}

}

### Years of Citing Articles

### OpenURL

### Abstract

This paper addresses the problem of segmenting an image into regions. We define a predicate for measuring the evidence for a boundary between two regions using a graph-based representation of the image. We then develop an e#cient segmentation algorithm based on this predicate, and show that although this algorithm makes greedy decisions it produces segmentations that satisfy global properties. We apply the algorithm to image segmentation using two different kinds of local neighborhoods in constructing the graph, and illustrate the results with both real and synthetic images. The algorithm runs in time nearly linear in the number of graph edges and is also fast in practice. An important characteristic of the method is its ability to preserve detail in low-variability image regions while ignoring detail in high-variability regions.

### Citations

8531 |
Introduction to Algorithms
- Cormen, Leiserson, et al.
- 2009
(Show Context)
Citation Context ...he one where every element is in its own component. We now turn to the segmentation algorithm, which is closely related to Kruskal’s algorithm for constructing a minimum spanning tree of a graph (cf. =-=[6]-=-). It can be implemented to run in O(m log m) time, where m is the number of edges in the graph. 10sAlgorithm 1 Segmentation algorithm. The input is a graph G = (V, E), with n vertices and m edges. Th... |

2590 | Normalized cuts and image segmentation
- Shi, Malik
- 1997
(Show Context)
Citation Context ...that runs at several frames per second can be used in video processing applications. While the past few years have seen considerable progress in eigenvector-based methods of image segmentation (e.g., =-=[14, 16]-=-), these methods are too slow to be practical for many applications. In contrast, the method described in this paper has been used in large-scale image database applications as described in [13]. Whil... |

2152 |
Algorithms for Clustering Data
- Jain, Dubes
- 1988
(Show Context)
Citation Context ...ding a good segmentation NP-hard. 2 Related Work There is a large literature on segmentation and clustering, dating back over 30 years, with applications in many areas other than computer vision (cf. =-=[9]-=-). In this section we briefly consider some of the related work that is most relevant to our approach: early graph-based methods (e.g., [15, 19]), region merging techniques (e.g., [5, 11]), techniques... |

319 | Segmentation using eigenvectors: A unifying view
- Weiss
- 1999
(Show Context)
Citation Context ...that runs at several frames per second can be used in video processing applications. While the past few years have seen considerable progress in eigenvector-based methods of image segmentation (e.g., =-=[14, 16]-=-), these methods are too slow to be practical for many applications. In contrast, the method described in this paper has been used in large-scale image database applications as described in [13]. Whil... |

268 | An optimal graph theoretic approach to data clustering: Theory and its application to image segmentation
- Wu, Leahy
- 1993
(Show Context)
Citation Context ...s (e.g., [15, 19]), region merging techniques (e.g., [5, 11]), techniques based on mapping image pixels to some feature space (e.g., [3, 4]) and more recent formulations in terms of graph cuts (e.g., =-=[14, 18]-=-) and spectral methods (e.g., [16]). Graph-based image segmentation techniques generally represent the problem in terms of a graph G = (V, E) where each node vi ∈ V corresponds to a pixel in the image... |

245 | Graph-theoretical methods for detecting and describing gestalt clusters
- Zahn
- 1971
(Show Context)
Citation Context ... and is computationally 2sefficient – running in O(n log n) time for n image pixels and with low constant factors, and can run in practice at video rates. As with certain classical clustering methods =-=[15, 19]-=-, our method is based on selecting edges from a graph, where each pixel corresponds to a node in the graph, and certain neighboring pixels are connected by undirected edges. Weights on each edge measu... |

231 |
Complexity and Approximation: Combinatorial Optimization Problems and Their Approximability Properties
- Ausiello, Crescenzi, et al.
- 1999
(Show Context)
Citation Context ...een the sets A and B. So it is the ratio of the number edges in E crossing the cut and the number of edges in F crossing the cut. Finding the value of the minimum ratio cut is an NP-hard problem (cf. =-=[2]-=-). First we show how to transform an instance of this problem to one where the sets E and F are disjoint, without modifying the value of the minimum cut. For every 21sedge (a, b) ∈ E ∩ F we create a n... |

187 | Robust analysis of feature spaces: color image segmentation
- Comaniciu, Meer
- 1997
(Show Context)
Citation Context ...ork that is most relevant to our approach: early graph-based methods (e.g., [15, 19]), region merging techniques (e.g., [5, 11]), techniques based on mapping image pixels to some feature space (e.g., =-=[3, 4]-=-) and more recent formulations in terms of graph cuts (e.g., [14, 18]) and spectral methods (e.g., [16]). Graph-based image segmentation techniques generally represent the problem in terms of a graph ... |

157 | Mean shift analysis and applications
- Comaniciu, Meer
- 1999
(Show Context)
Citation Context ...ork that is most relevant to our approach: early graph-based methods (e.g., [15, 19]), region merging techniques (e.g., [5, 11]), techniques based on mapping image pixels to some feature space (e.g., =-=[3, 4]-=-) and more recent formulations in terms of graph cuts (e.g., [14, 18]) and spectral methods (e.g., [16]). Graph-based image segmentation techniques generally represent the problem in terms of a graph ... |

148 | A factorization approach to grouping
- Perona, Freeman
- 1998
(Show Context)
Citation Context ...e of dissimilar color). For instance, this can result segmentations with regions that are disconnected in the image, which did not happen in the grid-graph case. Figure 6 shows a synthetic image from =-=[12]-=- and [8] and its segmentation, using k = 150 and with no smoothing (σ = 0). In this example the spatially disconnected regions do not reflect interesting scene structures, but we will see examples bel... |

100 | Image segmentation using local variation - Felzenszwalb, Huttenlocher - 1998 |

72 | Globally optimal regions and boundaries as minimum ratio weight cycles
- Jermyn, Ishikawa
(Show Context)
Citation Context ...n graphs. However, all such methods are too slow for many practical applications. An alternative to the graph cut approach is to look for cycles in a graph embedded in the image plane. For example in =-=[10]-=- the quality of each cycle is normalized in a way that is closely related to the normalized cuts approach. 3 Graph-Based Segmentation We take a graph-based approach to segmentation. Let G = (V, E) be ... |

40 | Graph theoretical clustering based on limited neighborhood sets - URQUHART - 1982 |

32 |
Approximate nearest neighbor searching
- Arya, Mount
- 1993
(Show Context)
Citation Context ...e pixels, and an overall running time of the segmentation method of O(n log n) time. There are many possible ways of picking a small fixed number of neighbors for each point. We use the ANN algorithm =-=[1]-=- to find the nearest neighbors for each point. This algorithm is quite fast in practice, given a 5-dimensional feature space with several hundred thousand points. The ANN method can also find approxim... |

27 |
A framework for learning query concepts in image classification
- Ratan, Maron
- 1999
(Show Context)
Citation Context ..., [14, 16]), these methods are too slow to be practical for many applications. In contrast, the method described in this paper has been used in large-scale image database applications as described in =-=[13]-=-. While there are other approaches to image segmentation that are highly efficient, these methods generally fail to capture perceptually important non-local properties of an image as discussed below. ... |

20 |
The tractability of segmentation and scene analysis
- Cooper
- 1998
(Show Context)
Citation Context ...uter vision (cf. [9]). In this section we briefly consider some of the related work that is most relevant to our approach: early graph-based methods (e.g., [15, 19]), region merging techniques (e.g., =-=[5, 11]-=-), techniques based on mapping image pixels to some feature space (e.g., [3, 4]) and more recent formulations in terms of graph cuts (e.g., [14, 18]) and spectral methods (e.g., [16]). Graph-based ima... |

13 |
of organization in perceptual forms (partial translation
- Laws
- 1938
(Show Context)
Citation Context ... organization, graph algorithm. 1 Introduction The problems of image segmentation and grouping remain great challenges for computer vision. Since the time of the Gestalt movement in psychology (e.g., =-=[17]-=-), it has been known that perceptual grouping plays a powerful role in human visual per1sception. A wide range of computational vision problems could in principle make good use of segmented images, we... |

3 |
Structural Pattern Recognition
- Pavlidas
- 1977
(Show Context)
Citation Context ...uter vision (cf. [9]). In this section we briefly consider some of the related work that is most relevant to our approach: early graph-based methods (e.g., [15, 19]), region merging techniques (e.g., =-=[5, 11]-=-), techniques based on mapping image pixels to some feature space (e.g., [3, 4]) and more recent formulations in terms of graph cuts (e.g., [14, 18]) and spectral methods (e.g., [16]). Graph-based ima... |

2 |
Stochastic clustering by typical cuts
- Gdalyahu, Weinshall, et al.
- 1999
(Show Context)
Citation Context ...imilar color). For instance, this can result segmentations with regions that are disconnected in the image, which did not happen in the grid-graph case. Figure 6 shows a synthetic image from [12] and =-=[8]-=- and its segmentation, using k = 150 and with no smoothing (σ = 0). In this example the spatially disconnected regions do not reflect interesting scene structures, but we will see examples below which... |