Unsupervised Image-to-Image Translation Networks

TL;DR

This paper introduces a novel unsupervised image-to-image translation framework using coupled GANs and a shared-latent space assumption, achieving high-quality translations across diverse domains and state-of-the-art domain adaptation results.

Contribution

The paper proposes a new unsupervised image translation method based on coupled GANs with a shared-latent space, improving translation quality and domain adaptation performance.

Findings

High-quality image translation across multiple domains

Outperforms existing methods in unsupervised translation tasks

Achieves state-of-the-art results in domain adaptation

Abstract

Unsupervised image-to-image translation aims at learning a joint distribution of images in different domains by using images from the marginal distributions in individual domains. Since there exists an infinite set of joint distributions that can arrive the given marginal distributions, one could infer nothing about the joint distribution from the marginal distributions without additional assumptions. To address the problem, we make a shared-latent space assumption and propose an unsupervised image-to-image translation framework based on Coupled GANs. We compare the proposed framework with competing approaches and present high quality image translation results on various challenging unsupervised image translation tasks, including street scene image translation, animal image translation, and face image translation. We also apply the proposed framework to domain adaptation and achieve state-of-the-art performance on benchmark datasets. Code and additional results are available in https://github.com/mingyuliutw/unit .