1. Shenlong Wang
  2. SCDL

Overview

HTTPS SSH

Semi-Coupled Dictionary Learning

This repo is the folder for training and testing semi-coupled dictionary learning for super-resolution. Author: Shenlong Wang, NPU/HK POLYU Latest Update: 6, June 2012

Citing

This code is for the paper: S. Wang, L. Zhang, Y. Liang and Q. Pan, "Semi-coupled Dictionary Learning with Applications in Super-resolution and Photo-Sketch Synthesis", in CVPR 2012. It has been tested on 64-bit LINUX.

Acknowledgements

Thanks Dr. Weisheng Dong for providing some useful MATLAB functions including clustering and nonlocal block matching. The sparse coding algorithm used in this package is from SPAMS toolbox [2]. Please make sure that the toolbox has been successfully installed in your computer.

  • collectPatches : Training patches collecting
  • Dict_Train.m : Semi-coupled dictionary learning
  • Image_SR.m : Image interpolation demo

Pre-trained dictionary

Please download the pre-trained dictionaries (SCDL_Data.zip) from the project website and copy the MAT file to the Data subfolder if you want to evaluate the super-resolution/interpolation performance.

References

Plese cite those paper below if you use this software:

[1] S. Wang, L. Zhang, Y. Liang and Q. Pan, "Semi-coupled Dictionary Learning with Applications in Super-resolution and Photo-Sketch Synthesis", in CVPR 2012.

[2] J. Mairal, F. Bach, J. Ponce and G. Sapiro, "Online Dictionary Learning for Sparse Coding", in ICML 2009

Contact:

  • {csslwang, cslzhang}@comp.polyu.edu.hk