Improving Style-Content Disentanglement in Image-to-Image Translation

Abstract

Unsupervised image-to-image translation methods have achieved tremendous success in recent years. However, it can be easily observed that their models contain significant entanglement which often hurts the translation performance. In this work, we propose a principled approach for improving style-content disentanglement in image-to-image translation. By considering the information flow into each of the representations, we introduce an additional loss term which serves as a content-bottleneck. We show that the results of our method are significantly more disentangled than those produced by current methods, while further improving the visual quality and translation diversity.

arXiv, 2020

Results

AFHQ

Content
Style
StarGAN2
    Ours     

StarGAN2
    Ours     

StarGAN2
    Ours     

AFHQ

Content
Style
StarGAN2
    Ours     

StarGAN2
    Ours     

StarGAN2
    Ours     

CUB

Content
Style
StarGAN2
    Ours     

StarGAN2
    Ours     

StarGAN2
    Ours     

BibTeX

@article{gabbay2020stylecontent,
  author    = {Aviv Gabbay and Yedid Hoshen},
  title     = {Improving Style-Content Disentanglement in Image-to-Image Translation},
  journal   = {arXiv preprint arXiv:2007.04964},
  year      = {2020}
}

References

  • [1] Choi et al. StarGAN v2: Diverse Image Synthesis for Multiple Domains. In CVPR, 2020.