Multi-domain Unsupervised Image-to-Image Translation with Appearance Adaptive Convolution

TL;DR

This paper introduces a multi-domain unsupervised image-to-image translation framework that uses appearance adaptive convolution and contrast learning to generate diverse, plausible images across multiple domains while preserving content.

Contribution

It presents a novel framework combining appearance adaptive convolution and contrast learning for effective multi-domain I2I translation with a single model.

Findings

Produces diverse and plausible images across multiple domains

Outperforms state-of-the-art methods in visual quality

Effectively disentangles content and appearance features

Abstract

Over the past few years, image-to-image (I2I) translation methods have been proposed to translate a given image into diverse outputs. Despite the impressive results, they mainly focus on the I2I translation between two domains, so the multi-domain I2I translation still remains a challenge. To address this problem, we propose a novel multi-domain unsupervised image-to-image translation (MDUIT) framework that leverages the decomposed content feature and appearance adaptive convolution to translate an image into a target appearance while preserving the given geometric content. We also exploit a contrast learning objective, which improves the disentanglement ability and effectively utilizes multi-domain image data in the training process by pairing the semantically similar images. This allows our method to learn the diverse mappings between multiple visual domains with only a single framework. We show that the proposed method produces visually diverse and plausible results in multiple domains compared to the state-of-the-art methods.