by Casper Hogenboom, Nils Simon, Lennart Lehmann
Project as part of the Deep Learning for Visual Recognition course at Aarhus University, Fall 2019.
See the Double-DIP-Dehaze jupyter notebook for a demo.
Based upon https://github.com/yossigandelsman/DoubleDIP
Original paper: http://www.wisdom.weizmann.ac.il/~vision/DoubleDIP/resources/DoubleDIP.pdf