Sample Code

MATLAB packages available below:
All code is Copyright 2007 University of Illinois. Clustering of multivariate mixed data

Matlab package: segmentation_code.zip

The usage is simple: if X is the data matrix (in column form), calling

Y = coding_seg(X, epsilon, false);

will segment the data so as to minimize the number of bits needed to

code the data vectors upto distortion epsilon^2. The last parameter

should be true if the data are known to be nonzero-mean. See "test.m"

for an example.

Contact: John Wright (jnwright@uiuc.edu).

Classification of multivariate data

Matlab package: MICL_code.zip

Contact: John Wright (jnwright@uiuc.edu)

Segmentation of natural images

Matlab package: tbes_1.0.zip

Installation: This texture segmentation algorithm uses the superpixel code of Mori et al., which includes some c files that need to be compiled. To do this open MATLAB and then run the following commands:

> cd superpixels/yu_imncut > mex csparse.c > mex ic.c > mex imnb.c > mex parmatV.c > mex spmd_imncut.cUsage: To test our texture segmentation algorithm, run the following command in MATLAB:

> test_texture_seg

Please refer to our image segmentation webpage for details and extensive results / benchmarking.

Contact: Shankar Rao (