Arbitrary Style Transfer with Structure Enhancement by Combining the Global and Local Loss

TL;DR

This paper proposes a novel style transfer method that enhances structure preservation by combining global and local loss functions, resulting in higher-quality stylized images that retain recognizable content structure.

Contribution

It introduces a new approach integrating local Lapstyle features and global depth information to improve structure retention in arbitrary style transfer.

Findings

Generated images show improved structural clarity and visual quality.

Outperforms existing methods on standard datasets.

Enhanced style transfer with better content preservation.

Abstract

Arbitrary style transfer generates an artistic image which combines the structure of a content image and the artistic style of the artwork by using only one trained network. The image representation used in this method contains content structure representation and the style patterns representation, which is usually the features representation of high-level in the pre-trained classification networks. However, the traditional classification networks were designed for classification which usually focus on high-level features and ignore other features. As the result, the stylized images distribute style elements evenly throughout the image and make the overall image structure unrecognizable. To solve this problem, we introduce a novel arbitrary style transfer method with structure enhancement by combining the global and local loss. The local structure details are represented by Lapstyle and the global structure is controlled by the image depth. Experimental results demonstrate that our method can generate higher-quality images with impressive visual effects on several common datasets, comparing with other state-of-the-art methods.