ParGAN: Learning Real Parametrizable Transformations

TL;DR

ParGAN introduces a flexible, controllable image-to-image translation framework that learns transformations with intuitive parameters, without needing paired data, enabling smooth interpolations and multiple transformations simultaneously.

Contribution

It generalizes cycle-consistent GANs to incorporate explicit transformation parameters, allowing for more controllable and versatile image transformations.

Findings

Learns transformations without paired data.

Supports multiple transformations simultaneously.

Enables smooth interpolation between transformations.

Abstract

Current methods for image-to-image translation produce compelling results, however, the applied transformation is difficult to control, since existing mechanisms are often limited and non-intuitive. We propose ParGAN, a generalization of the cycle-consistent GAN framework to learn image transformations with simple and intuitive controls. The proposed generator takes as input both an image and a parametrization of the transformation. We train this network to preserve the content of the input image while ensuring that the result is consistent with the given parametrization. Our approach does not require paired data and can learn transformations across several tasks and datasets. We show how, with disjoint image domains with no annotated parametrization, our framework can create smooth interpolations as well as learn multiple transformations simultaneously.