Learning Unsupervised Cross-domain Image-to-Image Translation Using a Shared Discriminator

TL;DR

This paper introduces a novel unsupervised cross-domain image translation method using a shared discriminator in GANs, achieving results comparable to attention-based models without added complexity.

Contribution

The paper proposes a shared discriminator approach for unsupervised image translation, simplifying the model while maintaining high-quality translation performance.

Findings

Achieves comparable quality to attention-based methods

Uses a single shared discriminator for two GANs

Effective in semantic-preserving image translation

Abstract

Unsupervised image-to-image translation is used to transform images from a source domain to generate images in a target domain without using source-target image pairs. Promising results have been obtained for this problem in an adversarial setting using two independent GANs and attention mechanisms. We propose a new method that uses a single shared discriminator between the two GANs, which improves the overall efficacy. We assess the qualitative and quantitative results on image transfiguration, a cross-domain translation task, in a setting where the target domain shares similar semantics to the source domain. Our results indicate that even without adding attention mechanisms, our method performs at par with attention-based methods and generates images of comparable quality.