Damien's Matlab library
This is a collection of Matlab functions I wrote over the years while working in the field of computer vision. Those seemed generic enough to be potentially useful to other people, so I made them available here. The package includes standalone demo scripts that test/demonstrate the most interesting functions of the library.
Please note that I can't provide any support or guarantees whatsoever on the quality of the code. Some functions use hard-coded parameters that were tweaked to particular applications, and the code may thus require some modifications to be useful. Also, if you reuse part of this library in your application, please acknowledge the author(s) mentionned in the original code.
A list of the functions is given below; the code of each function can be accessed individually, but since there are many dependencies within the library, I suggest downloading the whole library here, and unzipping its contents in a directory that you'll add to the Matlab path (with all its subdirectories).
/DmlMex/ MEX code (C code), mostly duplicates of other (Matlab) functions of the library.
/GaussianProcesses/ Regression with Gaussian processes; mostly refactored/documented code from www.gaussianprocess.org/gpml/.
- gp_getLikelihood.m - Gaussian likelihood for Gaussian processes.
- gp_makePrediction.m - Gaussian process prediction.
- gp_optimize.m - Minimize differentiable multivariate function using conjugate gradients.