image-to-image translation with conditional adversarial networks (pix2pix)

This is a method that can learn mappings from input to output images, enabling various image translation tasks.