Feature Statistics Mixing Regularization for Generative Adversarial Networks

Junho Kim, Yunjey Choi, Youngjung Uh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In generative adversarial networks, improving discriminators is one of the key components for generation performance. As image classifiers are biased toward texture and debiasing improves accuracy, we investigate 1) if the discriminators are biased, and 2) if debiasing the discriminators will improve generation performance. Indeed, we find empirical evidence that the discriminators are sensitive to the style (e.g., texture and color) of images. As a remedy, we propose feature statistics mixing regularization (FSMR) that encourages the discriminator's prediction to be invariant to the styles of input images. Specifically, we generate a mixed feature of an original and a reference image in the discriminator's feature space and we apply regularization so that the prediction for the mixed feature is consistent with the prediction for the original image. We conduct extensive experiments to demonstrate that our regularization leads to reduced sensitivity to style and consistently improves the performance of various GAN architectures on nine datasets. In addition, adding FSMR to recently-proposed augmentation-based GAN methods further improves image quality. Our code is available at https://github.com/naver-ai/FSMR.

Original languageEnglish
Title of host publicationProceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
PublisherIEEE Computer Society
Pages11284-11293
Number of pages10
ISBN (Electronic)9781665469463
DOIs
Publication statusPublished - 2022
Event2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022 - New Orleans, United States
Duration: 2022 Jun 192022 Jun 24

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2022-June
ISSN (Print)1063-6919

Conference

Conference2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
Country/TerritoryUnited States
CityNew Orleans
Period22/6/1922/6/24

Bibliographical note

Funding Information:
Acknowledgements The authors thank NAVER AI Lab researchers and Jun-Yan Zhu for constructive discussion. All experiments were conducted on NAVER Smart Machine Learning (NSML) platform [23, 32]. This work was partly supported by an IITP grant (No.2021-0-00155) and an NRF grant (NRF-2021R1G1A1095637). Both grants are funded by the Korean government (MSIT).

Publisher Copyright:
© 2022 IEEE.

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Feature Statistics Mixing Regularization for Generative Adversarial Networks'. Together they form a unique fingerprint.

Cite this