spectralMattingCode
所属分类：图形图像处理
开发工具：matlab
文件大小：14311KB
下载次数：23
上传日期：2012-09-23 21:41:15
上 传 者：
jijiao5403
说明： 这个程序主要是针对数字图像抠图的！用的是贝叶斯的方法！
(This program is mainly for digital image matting! Bayesian method!)
文件列表:
spectralMattingCode\calcMattingComponents.m (1812, 2007-06-11)
spectralMattingCode\compile_mincut.m (106, 2007-05-27)
spectralMattingCode\decideFB.m (300, 2007-04-29)
spectralMattingCode\downSmpIm.m (318, 2007-05-22)
spectralMattingCode\face.tif (137058, 2007-06-01)
spectralMattingCode\face_alpha.tif (11592, 2009-05-26)
spectralMattingCode\fastKmeans.m (2625, 2007-05-17)
spectralMattingCode\finalEnhancment.m (594, 2007-05-03)
spectralMattingCode\flat3DArray.m (296, 2007-06-06)
spectralMattingCode\getLaplacian1.m (1770, 2007-04-29)
spectralMattingCode\ind2bin.m (169, 2007-04-29)
spectralMattingCode\iterateProjComps.m (1065, 2007-06-06)
spectralMattingCode\kid.tif (134566, 2007-06-01)
spectralMattingCode\kid_alpha.tif (14678, 2007-06-11)
spectralMattingCode\kid_scribbles.tif (126890, 2007-06-01)
spectralMattingCode\kim.tif (458116, 2007-06-01)
spectralMattingCode\kim_alpha.tif (26970, 2007-06-07)
spectralMattingCode\kim_scribbles.tif (456554, 2007-06-01)
spectralMattingCode\maxflow-v2.2.src\adjacency_list\block.h (8191, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\adjacency_list\graph.cpp (2277, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\adjacency_list\graph.h (5794, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\adjacency_list\maxflow.cpp (11287, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\CHANGES.TXT (358, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\block.h (8191, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\graph.cpp (10770, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\graph.h (8467, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\Makefile (95, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\maxflow (27750, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\maxflow.cpp (20816, 2007-05-27)
spectralMattingCode\maxflow-v2.2.src\forward_star\maxflowFromFile.cpp (3822, 2007-06-07)
spectralMattingCode\maxflow-v2.2.src\LICENSE.TXT (18009, 2007-05-27)
spectralMattingCode\maxflow.dll (24576, 2007-06-07)
spectralMattingCode\maxflow.mexglx (28790, 2007-06-06)
spectralMattingCode\minCutPull.m (367, 2007-06-11)
spectralMattingCode\mymaxflow.cpp (4574, 2007-05-27)
spectralMattingCode\my_images.mat (28072040, 2007-06-07)
spectralMattingCode\runAll.asv (1452, 2009-06-15)
spectralMattingCode\runAll.m (1473, 2009-06-15)
... ...
###################################################################
# #
# Spectral Clustering - software for unsupervised computation #
# of matting components, and grouping these components to #
# a Foreground / Background alpha matte. #
# http://www.vision.huji.ac.il/SpectralMatting #
# #
# Anat Levin, Alex Rav-Acha, Dani Lischinski #
# 2007 #
# #
###################################################################
1. Introduction.
#################
This software library implements the spectral clustering algorithm
described in:
Spectral Matting
Anat Levin, Alex Rav-Acha and Dani Lischinski,
In IEEE Computer Society Conference on Computer Vision and
Pattern Recognition (CVPR), June 2007.
This software can be used freely for research purposes.
This code has been compiled and tasted using matlab7. We noticed
problems with earlier matlab versions.
For the components-grouping we wrapped to matlab the 'mincut' library of
Vladimir Kolmogorov (vnk@cs.cornell.edu) and Yuri Boykov (yuri@csd.uwo.ca),
described in
An Experimental Comparison of Min-Cut/Max-Flow Algorithms
for Energy Minimization in Vision.
Yuri Boykov and Vladimir Kolmogorov.
In IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI),
September 2004
(This software is also available at:
http://www.cs.cornell.edu/People/vnk/software.html)
Note: When a small number of clusters is used, the code can be
run without compiling the max-flow code.
2. How to use.
################
a. Compile the mincut library using 'compile_mincut'. As mentioned before,
this is not needed when using a small number of clusters.
b. To re-create some of the examples that were shown in the paper,
use 'runAll'.
c. To compute only alpha components (without grouping), one can use:
alpha_comps = SpectralMatting(prefix, downSmp, [], eigs_num, nclust);
近期下载者：
相关文件：
收藏者：