DIFFUSE INTERFACE MODELS ON GRAPHS FOR CLASSIFICATION OF HIGH DIMENSIONAL DATA

被引:118
作者
Bertozzi, Andrea L. [1 ]
Flenner, Arjuna [2 ]
机构
[1] Univ Calif Los Angeles, Dept Math, Los Angeles, CA 90095 USA
[2] USN, Air Weap Ctr, China Lake, CA USA
基金
美国国家科学基金会;
关键词
Nystrom extension; diffuse interfaces; image processing; high dimensional data; IMAGE; REGULARIZATION; DECONVOLUTION;
D O I
10.1137/11083109X
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
There are currently several communities working on algorithms for classification of high dimensional data. This work develops a class of variational algorithms that combine recent ideas from spectral methods on graphs with nonlinear edge/region detection methods traditionally used in the PDE-based imaging community. The algorithms are based on the Ginzburg-Landau functional which has classical PDE connections to total variation minimization. Convex-splitting algorithms allow us to quickly find minimizers of the proposed model and take advantage of fast spectral solvers of linear graph-theoretic problems. We present diverse computational examples involving both basic clustering and semisupervised learning for different applications. Case studies include feature identification in images, segmentation in social networks, and segmentation of shapes in high dimensional datasets.
引用
收藏
页码:1090 / 1118
页数:29
相关论文
共 57 条
[41]  
Mcadams A., 2010, P 2010 ACM SIGGRAPH, P65, DOI DOI 10.5555/1921427.1921438
[42]   MOTION OF MULTIPLE JUNCTIONS - A LEVEL SET APPROACH [J].
MERRIMAN, B ;
BENCE, JK ;
OSHER, SJ .
JOURNAL OF COMPUTATIONAL PHYSICS, 1994, 112 (02) :334-363
[43]  
Naeini M. M., 2007, P IAPR C MACH VIS AP
[44]  
Ratanamahatana C. A., 2002, P IEEE WORKSH DAT CL
[45]   Normalized cuts and image segmentation [J].
Shi, JB ;
Malik, J .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2000, 22 (08) :888-905
[46]   A simple min-cut algorithm [J].
Stoer, M ;
Wagner, F .
JOURNAL OF THE ACM, 1997, 44 (04) :585-591
[47]  
Szlam A, 2010, P 27 INT C MACH LEAR, P1039, DOI DOI 10.0RG/PAPERS/233.PDF
[48]  
Szlam A., 2009, 0968 UCLA CAM
[49]  
Szlam AD, 2008, J MACH LEARN RES, V9, P1711
[50]  
van de Weijer J, 2006, LECT NOTES COMPUT SC, V3952, P334