Unpaired Image Demoiréing Based on Cyclic Moiré Learning

We propose an end-to-end unsupervised learning approach to image demoiréing based on cyclic moiré learning. The proposed cyclic moiré learning consists of the moiré learning network and demoiréing network. The moiré learning network generates moiré images to construct a paired set of moiré and clean images. Then, the demoiréing network is trained using the generated paired dataset to remove moiré artifacts. Further, the moiré learning network and the demoiréing network are integrated together to be trained in an end-to-end manner. Experimental results demonstrate that the proposed algorithm outperforms state-of-the-art unsupervised image restoration al-gorithms.