Software for reproducibility
The SMALLbox, a test ground framework for sparse representation and dictionary learning.
Implementation of the K-SVD algorithm (training a dictionary to lead to sparse representation) along with some demonstration of its use for denoising.
MPTK provides a fast implementation of the Matching Pursuit algorithm for the sparse decomposition of multichannel signals. It comprises a C++ library, standalone command line utilities, and some scripts for running it and plotting the results through Matlab.
A new toolbox for new audio applications: inpaint audio like you inpaint images!
GAP (Greedy Analysis Pursuit) is a greedy algorithm to recover cosparse analysis signals from incomplete linear data. It may be loosely considered as the analysis counterpart of the Orthogonal Matching Pursuit algorithm.
GAPn is a version of GAP able to deal with noisy data.
For details on GAPn, see http://hal.inria.fr/hal-00691162
For details on GAP, see http://hal.inria.fr/inria-00602205
AOL (Analysis Operator Learning) is an alogrithm to learn analysis operator from noiseles data. It may be considered as the analysis couterpart of usual dictionary learning algorithms.
A version of AOL dealing with noise is under development.
For details on AOL : http://hal.inria.fr/inria-00583133/en
For details on a noise aware version of AOL : http://hal.inria.fr/hal-00661549/en
* Note that this software is automatically downloaded and installed when using the SMALLbox, so it does not need installing separately in this case.