Code for the paper:

[1] S. Georgoulis, V. Vanweddingen, M. Proesmans, L. Van Gool
"A Gaussian Process Latent Variable Model for BRDF Inference"
International Conference on Computer Vision. 2015 

Code requirements:
You need to have the 
1. ds-gplvm package (link: http://ibug.doc.ic.ac.uk/resources/ele_TIP15/),
2. gpml package (link: http://www.gaussianprocess.org/gpml/code/matlab/doc/index.html) and
3. minFunc optimization package (link: https://www.cs.ubc.ca/~schmidtm/Software/minFunc.html)
in your matlab path.
Find in zip files the original versions of these packages that were used in this paper.

Instructions:
1. Download MERL BRDF database (link: http://www.merl.com/brdf/).
2. Extract the zip files of the required matlab packages.
3. Copy and replace the modified functions from modified_functions folder to ds_gplvm folder.
4. In line 14 of my_dsgplvm.m file replace the path with the location of the binary files from MERL. 
5. Run the my_dsgplvm.m file.

Should you use the code, please cite [1].

Code released as is **for research purposes only** 
Feel free to modify/distribute but please cite [1]. 

contact: stam.georgoulis@esat.kuleuven.be