One of the best parts of CycleGAN is that it does not require paired input and output data. In many applications of style transfer, paired data is a critical piece of the training. CycleGAN was one of the first GAN implementations that was able to break that mold and reliably train models without paired input:
Unpaired input data from the CycleGAN paper
CycleGAN has fairly simply Convolutional Neural Networks (CNNs) for the generator and the discriminator. The real secret sauce to this particular paper is in the implementation of the architecture—how we stitch these networks together to train them to learn ...
