The original StyleGAN from NVIDIA allowing to mix and match facial features at multiple levels

Released in: A Style-Based Generator Architecture for Generative Adversarial Networks



The authors propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature. The new architecture leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images (e.g., freckles, hair), and it enables intuitive, scale-specific control of the synthesis. The new generator improves the state-of-the-art in terms of traditional distribution quality metrics, leads to demonstrably better interpolation properties, and also better disentangles the latent factors of variation. To quantify interpolation quality and disentanglement, the authors propose two new, automated methods that are applicable to any generator architecture. Finally, they introduce a new, highly varied and high-quality dataset of human faces.


Year Released

Key Links & Stats



A Style-Based Generator Architecture for Generative Adversarial Networks

@article{DBLP:journals/corr/abs-1812-04948, author = {Tero Karras and Samuli Laine and Timo Aila}, title = {A Style-Based Generator Architecture for Generative Adversarial Networks}, journal = {CoRR}, volume = {abs/1812.04948}, year = {2018}, url = {}, eprinttype = {arXiv}, eprint = {1812.04948}, timestamp = {Tue, 01 Jan 2019 15:01:25 +0100}, biburl = {}, bibsource = {dblp computer science bibliography,} }

ML Tasks

  1. General
  2. Image Generation
  3. Style Transfer

ML Platform

  1. Tensorflow


  1. Still Image


  1. General
  2. Facial
  3. Digital Human

CG Platform

  1. Not Applicable

Related organizations