Hi, Igor,How are you! I finally finished the code-writing job. Now the codes of T-SBL/T-MSBL are available and can be downloaded from the link:For reproducibility, it includes some demos which can re-produce the experiment results in my paper.I also post the code of tMFOCUSS (developed in my ICASSP 2011 paper) in my homepage. It can be downloaded from the link:Also, I wrote a new version of the MSBL code, which can be downloaded from here:Although our lab alumni, David, posted his code in his homepage, that code is not suitable for algorithm comparison in most cases (since the values of some parameters are not suitable). The new version has chosen better values for these parameters and added many comments and advices.I'll very appreciate if you can mention these codes via your blog and add them into the code collection.
Best regards,Zhilin
Zhilin noticed that the algorithms were to cumbersome to use with parameter tuning, so he spent the week-end rewriting part of the algorithms. Here is his new entry on the subject on his blog: New Versions of T-MSBL/T-SBL are Available for Download that features the new codes.
Thanks Zhilin , I'll feature your code soon on the reconstruction section of the Big Picture in Compressive Sensing.
Thanks Zhilin , I'll feature your code soon on the reconstruction section of the Big Picture in Compressive Sensing.
[1] Sparse Signal Recovery with Temporally Correlated Source Vectors Using Sparse Bayesian Learning by Zhilin Zhang, Bhaskar D. Rao.
[2] Z.Zhang, B.D.Rao, Iterative Reweighted Algorithms for Sparse Signal Recovery with Temporally Correlated Source Vectors, ICASSP 2011. Downloaded here
[3] Z.Zhang, B.D.Rao, Exploiting Correlation in Sparse Signal Recovery Problems: Multiple Measurement Vectors, Block Sparsity, and Time-Varying Sparsity.
Image Credit: NASA/JPL/Space Science Institute, N00171687.jpg was taken on May 21, 2011 and received on Earth May 22, 2011. The camera was pointing toward TITAN at approximately 2,313,374 kilometers away, and the image was taken using the CL1 and GRN filters.
No comments:
Post a Comment