From this and this questions on Super Resolution GAN, I don\'t understand to what model is mapped the high resolution image, and why is it passed as input of the combined mo
