Fetching the paper…

DRIT++: Diverse Image-to-Image Translation via Disentangled Representations · Around