U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation

TL;DR

U-GAT-IT introduces an unsupervised image-to-image translation method that uses attention and adaptive normalization to handle both holistic and shape-changing transformations, outperforming previous models.

Contribution

The paper presents a novel attention module and AdaLIN normalization for improved unsupervised image translation, capable of handling complex geometric changes.

Findings

Outperforms existing state-of-the-art models

Handles both shape and texture changes effectively

Demonstrates superior translation quality on various datasets

Abstract

We propose a novel method for unsupervised image-to-image translation, which incorporates a new attention module and a new learnable normalization function in an end-to-end manner. The attention module guides our model to focus on more important regions distinguishing between source and target domains based on the attention map obtained by the auxiliary classifier. Unlike previous attention-based method which cannot handle the geometric changes between domains, our model can translate both images requiring holistic changes and images requiring large shape changes. Moreover, our new AdaLIN (Adaptive Layer-Instance Normalization) function helps our attention-guided model to flexibly control the amount of change in shape and texture by learned parameters depending on datasets. Experimental results show the superiority of the proposed method compared to the existing state-of-the-art models with a fixed network architecture and hyper-parameters. Our code and datasets are available at https://github.com/taki0112/UGATIT or https://github.com/znxlwm/UGATIT-pytorch.