Resolution Dependent GAN Interpolation for Controllable Image Synthesis Between Domains
By Justin N. M. Pinkney, Doron Adler. 2020
154 citations recorded by the source corpus.
GANs can generate photo-realistic images from the domain of their training data. However, those wanting to use them for creative purposes often want to generate imagery from a truly novel domain, a task which GANs are inherently unable to do. It is also desirable to have a level of control so that there is a degree of artistic direction rather than purely curation of random results. Here we present a method for interpolating between generative models of the StyleGAN architecture in a resolution dependent manner. This allows us to generate images from an entirely novel domain and do this with a degree of control over the nature of the output.
Subfields: Generative Models
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Generative Models
- Daniel Cohen-Or
- Yuval Alaluf
- Rinon Gal
- Or Patashnik
- Eli Shechtman
- Dani Lischinski
- Amit H. Bermano
- Yotam Nitzan
Citations in the corpus
- StyleGAN-NADA (2021, methodology): “Some place restrictions on the space of trainable weights, either during training [25, 34] or by mixing the weights of the fine-tuned model with those of the original [31].”
- Cross-Domain and Disentangled Face Manipulation With 3D Guidance (2021, methodology): “Inspired by the recent observation in [11], [12], we perform latent-consistent finetuning that adapts the StyleGAN2 generator to another domain while keeping the latent spaces aligned.”
- Third Time's the Charm? Image and Video Editing with StyleGAN3 (2022, background): “Multiple works leverage the fine-tuning of StyleGAN2 generators for various applications, such as image-to-image translation [21,52,64,74,84] and inversion [7,10,56].”
- VOGUE: Try-On by StyleGAN Interpolation Optimization (2021, methodology): “Related to our method are also conditional GAN networks [23, 4], and GAN editing methods [25, 2, 6, 15, 8, 1].”
- StyleAlign: Analysis and Applications of Aligned StyleGAN Models (2021, methodology): “Some previous works (Pinkney & Adler, 2020; Kwong et al., 2021) that considered only similar domains have used the W / W + spaces.”
- HyperStyle: StyleGAN Inversion with HyperNetworks for Real Image Editing (2021, background): “, Toonify [51] and StyleGAN-NADA [18]).”