This page contains links to software for
Matching Pursuit (*)
decomposition of signals in stochastic time-frequency dictionaries of Gabor functions, written by Dobiesław Ircha under supervision of Piotr J. Durka, called hereafter "the authors".
Authors grant a nonexclusive license to use this software and documentation for education and research.
No part of the software or documentation can be included in any commercial product without prior obtaining a written permission of the authors.
This software is provided "as is" and without any express or implied warranties. We do not provide technical support.
This implementation is described in the paper Stochastic time-frequency dictionaries for Matching Pursuit, P.J. Durka, D. Ircha and K.J. Blinowska, IEEE Transactions on Signal Processing, Vol. 49, No. 3, pp. 507-510, March 2001
Before you download the software, you may try it in an interactive way here mp v.III (relatively stable & documented)
Finally, fast and experimental mp with Voronoy search written by D. Blacha (documentation only in Polish, sorry :(.
Matlab files for reading and post-processing results of MP decompositions stored in the binary format of the above packages: zip w/routines and demo
pthread.dll, available e.g. here, is needed to run mp31.exe and mp4.exe
Matching Pursuit (MP) is an iterative algorithm for adaptive approximations, first published under this name in Matching Pursuit with time-frequency dictionaries, S. Mallat and Z. Zhang, IEEE Transactions on Signal Processing, Dec 1993--software available at ftp://cs.nyu.edu/pub/wave/software/mpp.tar.Z. Similar approach was proposed in: Signal approximation via data-adaptive normalized Gaussian function and its applications for speech processing, S. Qian, D. Chen and K. Chen, Proc. ICASSP-92, San Francisco, CA, March 23-26, 1992, pp.141-144.