###################################################################
#                                                                 #
#    Jointly Optimized Regressors for Image Super-resolution      #
#                        (version 1.0)                            #
#  	Dengxin Dai, Radu Timofte, Luc Van Gool	                  #
#			   					  #
#      Copyright (c) 2015, ETH Zurich.  All rights reserved.      # 
###################################################################

Please cite the following paper, if you end up using this code 
 in your publication:

@InProceedings{JOR:EG15,
  author = {D. Dai and R. Timofte, and L. {Van Gool}},
  title = {Jointly Optimized Regressors for Image Super-resolution},
  booktitle = {Eurographs},
  year = {2015},
  keywords = {}
}
    
############### JOR Code #######################

1. This is a matlab code, and you may start with run_JOR.m.  

2. The code for training JOR is added, start with JOR_training.m. 
   The trained model will be saved into folder './models', or you
   can download the trained model from the project page 
   http://people.ee.ethz.ch/~daid/JOR/ and put them to folder './models'     

3. The code depends on vl-feat library. The library can be downloaded
   from the website at http://www.vlfeat.org/. For the sake of convenience,
   we have included vlfeat-0.9.16 into this code package. 

4. The code is built on top of Radu Timofte's code for their work A+. Also, 
   the code of SRCNN by Chao Dong and colleages are also included for comparison.   

5. Feel free to contact us if you have any questions about using the code. 


Dengxin Dai, dai@vision.ee.ethz.ch, ddx2004@gmail.com