Improving Shape Deformation in Unsupervised Image-to-Image Translation

Aaron Gokaslan, Vivek Ramanujan, Daniel Ritchie, Kwang In Kim, James Tompkin

ECCV 2018

doi:10.1007/978-3-030-01258-8_40 /eccv/2018/gokaslan2018eccv-improving/

Abstract

Unsupervised image-to-image translation techniques are able to map local texture between two domains, but they are typically un- successful when the domains require larger shape change. Inspired by semantic segmentation, we introduce a discriminator with dilated convo- lutions which is able to use information from across the entire image to train a more context-aware generator. This is coupled with a multi-scale perceptual loss which is better able to represent error in the underly- ing shape of objects. We demonstrate that this design is more capable of representing shape deformation in a challenging toy dataset, plus in complex mappings with significant dataset variation between humans, dolls, and anime faces, and between cats and dogs.

PDF ECCV Semantic Scholar

Cite

Text

Gokaslan et al. "Improving Shape Deformation in Unsupervised Image-to-Image Translation." Proceedings of the European Conference on Computer Vision (ECCV), 2018. doi:10.1007/978-3-030-01258-8_40

Markdown

[Gokaslan et al. "Improving Shape Deformation in Unsupervised Image-to-Image Translation." Proceedings of the European Conference on Computer Vision (ECCV), 2018.](https://mlanthology.org/eccv/2018/gokaslan2018eccv-improving/) doi:10.1007/978-3-030-01258-8_40

BibTeX

@inproceedings{gokaslan2018eccv-improving,
  title     = {{Improving Shape Deformation in Unsupervised Image-to-Image Translation}},
  author    = {Gokaslan, Aaron and Ramanujan, Vivek and Ritchie, Daniel and In Kim, Kwang and Tompkin, James},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2018},
  doi       = {10.1007/978-3-030-01258-8_40},
  url       = {https://mlanthology.org/eccv/2018/gokaslan2018eccv-improving/}
}