### Citations

3787 | Normalized cuts and image segmentation.
- Shi, Malik
- 2000
(Show Context)
Citation Context ... effectiveness of DACA. 1 Introduction In recent years, spectral clustering (SC) has been successfully applied to many domains such as circuit layout [1, 2], load balancing [3] and image segmentation =-=[4, 5]-=-. Based on local evidence from similarities among data points, SC finds out the best graph cuts by optimizing a particular partitioning criterion function through eigendecomposition. With effectivenes... |

1708 | On Spectral Clustering: Analysis and an Algorithm
- Ng, Jordan, et al.
- 2001
(Show Context)
Citation Context ... algorithm for graph partitioning. Ding et al. [7] present a min-max cut algorithm for graph partitioning and data clustering. Balanced partitions are obtained by the min-max cut algorithm. Ng et al. =-=[8]-=- present a clustering algorithm based on KMeans after the spectral relaxation. Yu and Shi [9] propose a principled account on multiclass spectral clustering. They give a nearly global-optimal discrete... |

404 | Contour and texture analysis for image segmentation
- Malik, Belongie, et al.
- 2001
(Show Context)
Citation Context ... effectiveness of DACA. 1 Introduction In recent years, spectral clustering (SC) has been successfully applied to many domains such as circuit layout [1, 2], load balancing [3] and image segmentation =-=[4, 5]-=-. Based on local evidence from similarities among data points, SC finds out the best graph cuts by optimizing a particular partitioning criterion function through eigendecomposition. With effectivenes... |

265 | Multiclass spectral clustering.
- Yu, Shi
- 2003
(Show Context)
Citation Context ...partitioning and data clustering. Balanced partitions are obtained by the min-max cut algorithm. Ng et al. [8] present a clustering algorithm based on KMeans after the spectral relaxation. Yu and Shi =-=[9]-=- propose a principled account on multiclass spectral clustering. They give a nearly global-optimal discrete clustering solution by using singular value decomposition and nonmaximum suppression in an i... |

220 |
An improved spectral graph partitioning algorithm for mapping parallel computations
- Hendrickson, Leland
- 1992
(Show Context)
Citation Context ...demonstrate the promise and effectiveness of DACA. 1 Introduction In recent years, spectral clustering (SC) has been successfully applied to many domains such as circuit layout [1, 2], load balancing =-=[3]-=- and image segmentation [4, 5]. Based on local evidence from similarities among data points, SC finds out the best graph cuts by optimizing a particular partitioning criterion function through eigende... |

213 | A Min-max cut algorithm for graph partitioning and data clustering
- Ding, He, et al.
- 2001
(Show Context)
Citation Context ...class data learning. Shi and Malik [4] propose a normalized cut criterion for segmenting the similarity graph. Gdalyahu et al.[6] present a “typical cut” algorithm for graph partitioning. Ding et al. =-=[7]-=- present a min-max cut algorithm for graph partitioning and data clustering. Balanced partitions are obtained by the min-max cut algorithm. Ng et al. [8] present a clustering algorithm based on KMeans... |

173 |
Spectral k-way ratio-cut partitioning and clustering,”
- Chan, Schlag, et al.
- 1993
(Show Context)
Citation Context ...perimental evaluations demonstrate the promise and effectiveness of DACA. 1 Introduction In recent years, spectral clustering (SC) has been successfully applied to many domains such as circuit layout =-=[1, 2]-=-, load balancing [3] and image segmentation [4, 5]. Based on local evidence from similarities among data points, SC finds out the best graph cuts by optimizing a particular partitioning criterion func... |

89 | Self-Organization in Vision: Stochastic Clustering for Image Segmentation, Perceptual Grouping, and Image Database Organization,”
- Gdalyahu, Weinshall, et al.
- 2001
(Show Context)
Citation Context ...iveness in clustering data of complex structure, SC is promising for multiclass data learning. Shi and Malik [4] propose a normalized cut criterion for segmenting the similarity graph. Gdalyahu et al.=-=[6]-=- present a “typical cut” algorithm for graph partitioning. Ding et al. [7] present a min-max cut algorithm for graph partitioning and data clustering. Balanced partitions are obtained by the min-max c... |

33 | Learning activity patterns using fuzzy self-organizing neural network - Hu, Xie, et al. - 2004 |

21 | Multiway partitioning via geometric embeddings, orderings, and dynamic programming.
- Alpert, Kahng
- 1995
(Show Context)
Citation Context ...perimental evaluations demonstrate the promise and effectiveness of DACA. 1 Introduction In recent years, spectral clustering (SC) has been successfully applied to many domains such as circuit layout =-=[1, 2]-=-, load balancing [3] and image segmentation [4, 5]. Based on local evidence from similarities among data points, SC finds out the best graph cuts by optimizing a particular partitioning criterion func... |

14 |
Motion Trajectory Learning in the DFTCoefficient Feature Space,”
- Naftel, Khalid
- 2006
(Show Context)
Citation Context ...ng the nodes in the graph. The last experiment on the trajectory dataset is performed to showcase the performance of DACA, The final clustering results are shown in Fig. 5. The DFTcoefficient feature =-=[11]-=- is used again to represent the trajectories. Raw trajectories are clustered into nine trajectory classes. In each trajectory class, trajectories have very similar directions. We just choose to displa... |

8 |
A multi-object tracking system for surveillance video analysis
- Xie, Hu, et al.
(Show Context)
Citation Context ...es. The first class is an outlier while the other three are dominant. The last dataset is formed by a total number of 1200 vehicle motion trajectories, which is acquired from the tracker presented in =-=[12]-=-. k in (4) is set as 5. Two experiments are conducted to demonstrate the claimed contributions of the proposed DAC-based clustering algorithm (DACA). They are to evaluate the clustering accuracy of DA... |

4 | Nonlinear multiclass discriminant analysis,
- Ma, Sancho-Gomez, et al.
- 2003
(Show Context)
Citation Context ...stering results of DACA. maximize h(P ) = 1K tr{(PTQP )−1(PT ŴP )} subject to PTP = IK . (7) The optimization problem (7) has been addressed in multiclass LDA (linear discriminant analysis) learning =-=[13]-=-. A solution P̃ to (7) consists of the K principal eigenvectors (i.e., corresponding to the K largest eigenvalues) of the matrix Q−1Ŵ . If Q is a singular matrix, Q−1Ŵ should be replaced with the ma... |