Implicit pairs for boosting unpaired image-to-image translation  

作　　者：Yiftach Ginger Dov Danon Hadar Averbuch-Elor Daniel Cohen-Or 

机构地区：[1]Tel Aviv University,Israel

出　　处：《Visual Informatics》2020年第4期50-58,共9页可视信息学（英文）

摘　　要：In image-to-image translation the goal is to learn a mapping from one image domain to another.In the case of supervised approaches the mapping is learned from paired samples.However,collecting large sets of image pairs is often either prohibitively expensive or not possible.As a result,in recent years more attention has been given to techniques that learn the mapping from unpaired sets.In our work,we show that injecting implicit pairs into unpaired sets strengthens the mapping between the two domains,improves the compatibility of their distributions,and leads to performance boosting of unsupervised techniques by up to 12%across several measurements.The competence of the implicit pairs is further displayed with the use of pseudo-pairs,i.e.,paired samples which only approximate a real pair.We demonstrate the effect of the approximated implicit samples on image-to-image translation problems,where such pseudo-pairs may be synthesized in one direction,but not in the other.We further show that pseudo-pairs are significantly more effective as implicit pairs in an unpaired setting,than directly using them explicitly in a paired setting.

关 键 词：Generative adversarial networks Image-to-image translation Data augmentation Synthetic samples 

分 类 号：H31[语言文字—英语]