Pixel-Level Domain Transfer
This work addresses image-to-image translation for domain transfer, but it is incremental as it builds on existing GAN methods with a novel discriminator.
The paper tackles the problem of generating realistic target domain images from input images by introducing a domain-discriminator alongside a GAN framework, achieving decent results in a clothing generation task.
We present an image-conditional image generation model. The model transfers an input domain to a target domain in semantic level, and generates the target image in pixel level. To generate realistic target images, we employ the real/fake-discriminator as in Generative Adversarial Nets, but also introduce a novel domain-discriminator to make the generated image relevant to the input image. We verify our model through a challenging task of generating a piece of clothing from an input image of a dressed person. We present a high quality clothing dataset containing the two domains, and succeed in demonstrating decent results.