Style Transfer by Rigid Alignment in Neural Net Feature Space

TL;DR

This paper introduces a novel style transfer method that aligns style features to content features using rigid alignment, resulting in high-quality stylized images that preserve content structure more effectively than existing methods.

Contribution

The paper proposes a new approach for arbitrary style transfer that reverses the typical feature transformation process by aligning style features to content features with rigid alignment.

Findings

Produces high-quality stylized images

Preserves content structure better than previous methods

Outperforms current state-of-the-art techniques in style transfer

Abstract

Arbitrary style transfer is an important problem in computer vision that aims to transfer style patterns from an arbitrary style image to a given content image. However, current methods either rely on slow iterative optimization or fast pre-determined feature transformation, but at the cost of compromised visual quality of the styled image; especially, distorted content structure. In this work, we present an effective and efficient approach for arbitrary style transfer that seamlessly transfers style patterns as well as keep content structure intact in the styled image. We achieve this by aligning style features to content features using rigid alignment; thus modifying style features, unlike the existing methods that do the opposite. We demonstrate the effectiveness of the proposed approach by generating high-quality stylized images and compare the results with the current state-of-the-art techniques for arbitrary style transfer.