arXiv:1808.04325cs.CV
Improving Shape Deformation in Unsupervised Image-to-Image Translation
Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.
Abstract
Unsupervised image-to-image translation techniques are able to map local texture between two domains, but they are typically unsuccessful when the domains require larger shape change. Inspired by semantic segmentation, we introduce a discriminator with dilated convolutions that is able to use information from across the entire image to train a more context-aware generator. This is coupled with a multi-scale perceptual loss that is better able to represent error in the underlying shape of objects. We demonstrate that this design is more capable of representing shape deformation in a challenging toy dataset, plus in complex mappings with significant dataset variation between humans, dolls, and anime faces, and between cats and dogs.
(I contributed that one.) It's kind of annoying because when you first see CycleGAN, you think, 'this is amazing! I can use it for anything, like turning faces<->anime!' And then you actually try and you realize your photo<->anime CycleGAN is just learning the steganographic…
gwernView on HNParent context