Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data
This addresses a limitation in structured prediction tasks like image segmentation by enabling flexible mappings without paired data, though it appears incremental as an extension of CycleGAN.
The paper tackles the problem of learning many-to-many mappings between domains from unpaired data, which CycleGAN fails to handle due to its one-to-one assumption, and demonstrates the effectiveness of Augmented CycleGAN on image datasets.
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.