In multimodal unsupervised image-to-image translation tasks, the goal is to
translate an image from the source domain to many images in the target domain.
We present a simple method that produces higher quality images than current
state-of-the-art while maintaining the same amount of m