textureSynth

This package contains MatLab code for analyzing and synthesizing digital image of visual texture. The algorithm is described in the references given at the bottom of this document. Further information, as well the most recent versions of the code, are available at

http://www.cns.nyu.edu/~lcv/texture/

Incremental changes to the code are documented in the ChangeLog file.

Written by Javier Portilla and Eero Simoncelli, 1999-2000.

Comments/Suggestions/Bugs to:

• javier@decsai.ugr.es
• eero.simoncelli@nyu.edu

Tip

An actively-maintained GPU-compatible python port of this model (using pytorch) can be found in the plenoptic package. Note that the plenoptic port is not exactly identical: it makes use of pytorch's built-in optimization, rather than the custom optimization found here, and it returns only the statistics described in the paper (this implementation includes some redundant statistics, see plenoptic docs for more details). The results of texture synthesis are still qualitatively similar.

INSTALLATION

1. download and unpack the code. You can put the code anywhere on your system, but we'll assume it's in a directory named textureSynth.

2. download and unpack the matlabPyrTools package This is a collection of tools for multi-scale decomposition of images. You can put the code anywhere on your system, but we'll assume it's in a directory named matlabPyrTools. Please use version 1.4 or newer of the matlabPyrTools.

3. Run matlab, and put the matlabPyrTools directory in your path: > path('matlabPyrTools', path);

4. The matlabPyrTools distribution includes a MEX subdirectory containing binary executables, precompiled for various platforms (SunOS,Solaris, Linux,Windows). You may need to recompile these on your platform. In addition, you should either move the relavent files from the MEX subdirectory into the main directory, OR create a link/alias to them, OR place the MEX subdirectory in your matlab path.

USING THE SOFTWARE

• To see a demonstration, start Matlab, change directories (using cd) into the textureSynth directory, and execute example1.

• If you want to learn how to use the texture analysis and synthesis functions, take a look at example1.m and example2.m (these include many explanatory comments).

• For a listing of matlab function included in this package, execute help textureSynth

• For details on any of the functions, execute help <name_of_function>

REFERENCES

J Portilla and E P Simoncelli. A Parametric Texture Model based on Joint Statistics of Complex Wavelet Coefficients. Int'l Journal of Computer Vision. 40(1):49-71, October, 2000. http://www.cns.nyu.edu/~eero/ABSTRACTS/portilla99-abstract.html

J Portilla and E P Simoncelli Texture Modeling and Synthesis using Joint Statistics of Complex Wavelet Coefficients. IEEE Workshop on Statistical and Computational Theories of Vision, Fort Collins, CO, 22 June 1999. http://www.cns.nyu.edu/~eero/ABSTRACTS/portilla99a-abstract.html

J Portilla and E P Simoncelli. Texture Representation and Synthesis Using Correlation of Complex Wavelet Coefficient Magnitudes. Technical Report #54, Consejo Superior de Investigaciones Cientificas (CSIC), Madrid. 29 March 1999. http://www.cns.nyu.edu/~eero/ABSTRACTS/portilla98-abstract.html

E P Simoncelli and J Portilla. Texture Characterization via Joint Statistics of Wavelet Coefficient Magnitudes. In 5th IEEE Int'l Conf on Image Processing. Chicago, IL. Oct 4-7, 1998. http://www.cns.nyu.edu/~eero/ABSTRACTS/simoncelli98b-abstract.html