Breaking the cycle -- Colleagues are all you need

TL;DR

This paper introduces a novel unpaired image-to-image translation method that leverages collaboration between multiple GANs, producing diverse outputs without cycle constraints, and effectively handling large shape modifications and object removal.

Contribution

It proposes a multi-GAN collaboration approach for unpaired image translation that avoids cycle constraints and improves translation quality and diversity.

Findings

Outperforms state-of-the-art methods on challenging datasets

Removes large objects like glasses effectively

Produces diverse and high-quality translated images

Abstract

This paper proposes a novel approach to performing image-to-image translation between unpaired domains. Rather than relying on a cycle constraint, our method takes advantage of collaboration between various GANs. This results in a multi-modal method, in which multiple optional and diverse images are produced for a given image. Our model addresses some of the shortcomings of classical GANs: (1) It is able to remove large objects, such as glasses. (2) Since it does not need to support the cycle constraint, no irrelevant traces of the input are left on the generated image. (3) It manages to translate between domains that require large shape modifications. Our results are shown to outperform those generated by state-of-the-art methods for several challenging applications on commonly-used datasets, both qualitatively and quantitatively.