Leveraging in-domain supervision for unsupervised image-to-image translation tasks via multi-stream generators

Open in new window