ISF-GAN: An Implicit Style Function for High-Resolution Image-to-Image Translation

TL;DR

This paper introduces ISF-GAN, a novel implicit style function that enables high-resolution, multi-modal, and multi-domain image-to-image translation using pre-trained unconditional GANs, improving over existing methods.

Contribution

The paper proposes ISF-GAN, a new method for multi-modal, multi-domain image translation that preserves identity and leverages pre-trained GANs for high-resolution outputs.

Findings

Significantly improved results over baselines in face and animal image translation.

Enables cost-effective high-resolution multi-modal image translation.

Supports multiple visual domains and preserves domain-invariant features.

Abstract

Recently, there has been an increasing interest in image editing methods that employ pre-trained unconditional image generators (e.g., StyleGAN). However, applying these methods to translate images to multiple visual domains remains challenging. Existing works do not often preserve the domain-invariant part of the image (e.g., the identity in human face translations), they do not usually handle multiple domains, or do not allow for multi-modal translations. This work proposes an implicit style function (ISF) to straightforwardly achieve multi-modal and multi-domain image-to-image translation from pre-trained unconditional generators. The ISF manipulates the semantics of an input latent code to make the image generated from it lying in the desired visual domain. Our results in human face and animal manipulations show significantly improved results over the baselines. Our model enables cost-effective multi-modal unsupervised image-to-image translations at high resolution using pre-trained unconditional GANs. The code and data are available at: \url{https://github.com/yhlleo/stylegan-mmuit}.