Authors:
- Rosa M. Figueras i Ventura
- Oscar Divorra Escoda

Source code for the MP image coder presented in my PhD and in the IEEE TIP paper 
"Low rate and scalable image coding with redundant representations". 
This code only provides the MP decomposition code (not the arithmetic encoder or the quantizer).

Disclaimer:

* There is a copyright on this software, please cite these publications if you use it:

- Rosa M. Figueras i Ventura, Oscar Divorra Escoda and Pierre Vandergheynst, 
A Matching Pursuit Full Search Algorithm for Image Approximations, 
ITS Technical Report TR-ITS-31/2004, December 2004

- Rosa M. Figueras i Ventura, Pierre Vandergheynst and Pascal Frossard, 
Low rate and scalable image coding with redundant representations. I
EEE Transactions on Image Processing, Volume 15, Issue 3, March 2006 Page(s):726 - 739 
 
* Not available for commercial use.

* Use it under your own risk. 


Usage (Unix): 

- Untar file MPCoder.tgz (tar -xvzf MPCoder.tgz)
- Place yourself in the folder MP2D
- Type make (and the compilation options you wish to use)
- The coder should compile, the executable files are placed in MP2D/bin
- Use MP2D_3.0 (found in MP2D/bin/3.0). The coder is "testMPClass" 

For Windows or other operating system users: sorry, the makefile is only for Unix (Linux) users. Use Cygwin. Good luck...