Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data

[edit]

Amjad Almahairi, Sai Rajeshwar, Alessandro Sordoni, Philip Bachman, Aaron Courville ;
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:195-204, 2018.

Abstract

Learning inter-domain mappings from unpaired data can improve performance in structured prediction tasks, such as image segmentation, by reducing the need for paired data. CycleGAN was recently proposed for this problem, but critically assumes the underlying inter-domain mapping is approximately deterministic and one-to-one. This assumption renders the model ineffective for tasks requiring flexible, many-to-many mappings. We propose a new model, called Augmented CycleGAN, which learns many-to-many mappings between domains. We examine Augmented CycleGAN qualitatively and quantitatively on several image datasets.

Related Material