Adaptive sparse representation based on visual saliency for image compression.For salient region,it gives more sparsity level (0-norm) to further improve performance.This code is a simple implementation of paper:
Image compression using adaptive sparse representations over trained dictionaries
I'm not ensure the correctness of this implementation,and the dictionary isn't trained as described in the paper.So please use carefully.
To use this code,you should unzip the gbvs.rar file first and install gbvs,more details in http://www.klab.caltech.edu/~harel/share/gbvs.php
If you have any questions or find mistakes on this code,please contact my:[email protected]