Corpus ID: 211259173

GANHopper: Multi-Hop GAN for Unsupervised Image-to-Image Translation

@article{Lira2020GANHopperMG,
  title={GANHopper: Multi-Hop GAN for Unsupervised Image-to-Image Translation},
  author={Wallace P. Lira and Johannes Merz and Daniel Ritchie and Daniel Cohen-Or and Hao Zhang},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.10102}
}
  • Wallace P. Lira, Johannes Merz, +2 authors Hao Zhang
  • Published 2020
  • Computer Science
  • ArXiv
  • We introduce GANHopper, an unsupervised image-to-image translation network that transforms images gradually between two domains, through multiple hops. Instead of executing translation directly, we steer the translation by requiring the network to produce in-between images which resemble weighted hybrids between images from the two in-put domains. Our network is trained on unpaired images from the two domains only, without any in-between images.All hops are produced using a single generator… CONTINUE READING

    Figures and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 27 REFERENCES

    DualGAN: Unsupervised Dual Learning for Image-to-Image Translation

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    TransGaGa: Geometry-Aware Unsupervised Image-To-Image Translation

    VIEW 3 EXCERPTS

    CariGANs: unpaired photo-to-caricature translation

    VIEW 1 EXCERPT