Iconify: Converting Photographs into Icons
This addresses a domain conversion problem for graphic design or image processing applications, but it is incremental as it applies existing GAN methods to a new task.
The paper tackled converting photographs to icons using generative adversarial networks (GANs) without one-to-one correspondence, and found that CycleGAN learned sufficient abstraction to generate icon-like images in experiments.
In this paper, we tackle a challenging domain conversion task between photo and icon images. Although icons often originate from real object images (i.e., photographs), severe abstractions and simplifications are applied to generate icon images by professional graphic designers. Moreover, there is no one-to-one correspondence between the two domains, for this reason we cannot use it as the ground-truth for learning a direct conversion function. Since generative adversarial networks (GAN) can undertake the problem of domain conversion without any correspondence, we test CycleGAN and UNIT to generate icons from objects segmented from photo images. Our experiments with several image datasets prove that CycleGAN learns sufficient abstraction and simplification ability to generate icon-like images.