Rethink Arbitrary Style Transfer with Transformer and Contrastive Learning

TL;DR

This paper introduces a novel style transfer method using a perception encoder, style consistency normalization, and contrastive learning to produce higher quality stylized images with fewer artifacts.

Contribution

It proposes a new framework combining SCIN, ICL, and PE to improve style transfer quality, addressing limitations of previous cross-attention and adaptive normalization methods.

Findings

Produces higher quality stylized images with fewer artifacts

Outperforms existing state-of-the-art methods in experiments

Enhances style-feature understanding through contrastive learning

Abstract

Arbitrary style transfer holds widespread attention in research and boasts numerous practical applications. The existing methods, which either employ cross-attention to incorporate deep style attributes into content attributes or use adaptive normalization to adjust content features, fail to generate high-quality stylized images. In this paper, we introduce an innovative technique to improve the quality of stylized images. Firstly, we propose Style Consistency Instance Normalization (SCIN), a method to refine the alignment between content and style features. In addition, we have developed an Instance-based Contrastive Learning (ICL) approach designed to understand the relationships among various styles, thereby enhancing the quality of the resulting stylized images. Recognizing that VGG networks are more adept at extracting classification features and need to be better suited for capturing style features, we have also introduced the Perception Encoder (PE) to capture style features. Extensive experiments demonstrate that our proposed method generates high-quality stylized images and effectively prevents artifacts compared with the existing state-of-the-art methods.