%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2019/09.16.01.10
%2 sid.inpe.br/sibgrapi/2019/09.16.01.10.43
%@doi 10.1109/SIBGRAPI.2019.00025
%T Exploring Double Cross Cyclic Interpolation in Unpaired Image-to-Image Translation
%D 2019
%A Cáceres, Jorge Roberto López,
%A Condori, Manasses Antoni Mauricio,
%A Chávez, Guillermo Cámara,
%@affiliation Universidad Católica San Pablo
%@affiliation Universidad Católica San Pablo
%@affiliation Federal University of Ouro Preto
%E Oliveira, Luciano Rebouças de,
%E Sarder, Pinaki,
%E Lage, Marcos,
%E Sadlo, Filip,
%B Conference on Graphics, Patterns and Images, 32 (SIBGRAPI)
%C Rio de Janeiro, RJ, Brazil
%8 28-31 Oct. 2019
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K Unpaired Image-to-Image Translation, Latent Space Interpolation, Cross-domain Model.
%X The unpaired image-to-image translation consists of transferring a sample $a$ in the domain $A$ to an analog sample $b$ in the domain $B$ without intensive pixel-to-pixel supervision. The current vision focuses on learning a generative function that maps both domains but ignoring the latent information, although its exploration is not explicit supervision. This paper proposes a cross-domain GAN-based model to achieve a bi-directional translation guided by latent space supervision. The proposed architecture provides a double-loop cyclic reconstruction loss in an exchangeable training adopted to reduce mode collapse and enhance local details. Our proposal has outstanding results in visual quality, stability, and pixel-level segmentation metrics over different public datasets.
%@language en
%3 Sibgrapi19_CycleGAN.pdf