hardmaru: GANILLA: Generative Adversarial Networks for Image to Illustration Translation
code/pre-trained models/dataset https://github.com/giddyyupp/ganilla
Style transfer using Ghibli anime: https://t.co/fk2uWIO1k8
5 replies, 540 likes
👩💻 DynamicWebPaige: "In this paper, we explore illustrations in children’s books as a new domain in unpaired image-to-image translation."
😄 Too cool!
Note to self: create pictures of nieces in the illustration style of Arthur, Chrysanthemum, and Where the Wild Things Are. https://t.co/F0NBmzr4nQ
1 replies, 128 likes
✨ Joe Fabisevich™ ✨: This is one of the coolest uses of machine learning ever. I don't care if it's not useful, but the idea of turning everything into Studio Ghibli art makes me really happy.
2 replies, 100 likes
roadrunner01: GANILLA: Generative Adversarial Networks for Image to
github: https://github.com/giddyyupp/ganilla https://t.co/UUimL70wVA
0 replies, 94 likes
Julianne Dalcanton: “We love illustration so let’s make this nifty thing that will lead to cutting back one of the few ways artists get paid.” #UnintendedConsequences
0 replies, 2 likes
Found on Feb 16 2020 at https://arxiv.org/pdf/2002.05638.pdf