Large Scale GAN Training for High Fidelity Natural Image Synthesis

By Andrew Brock, Jeff Donahue, Karen Simonyan. 2018

6,195 citations recorded by the source corpus.

Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple "truncation trick," allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, 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 Computer Graphics and Computer-Aided Design

  1. Andreas Geiger
  2. Eli Shechtman
  3. Daniel Cohen-Or
  4. Michael Niemeyer
  5. Jonathan T. Barron
  6. Alexei A. Efros
  7. Jun-Yan Zhu
  8. Richard Zhang

In Computer Vision and Pattern Recognition

  1. Luc Van Gool
  2. Dacheng Tao
  3. Ming-Hsuan Yang
  4. Trevor Darrell
  5. Xiaogang Wang
  6. Jitendra Malik
  7. Andrew Zisserman
  8. Andrea Vedaldi

In Generative Models

  1. Jun-Yan Zhu
  2. Eli Shechtman
  3. Stefano Ermon
  4. Richard Zhang
  5. Arash Vahdat
  6. Daniel Cohen-Or
  7. Alexei A. Efros
  8. Jiaming Song

Citations in the corpus