Neural Comic Style Transfer: Case Study

TL;DR

This paper compares various neural style transfer methods for applying comic styles to images, analyzing their strengths and weaknesses through qualitative, quantitative, and user survey evaluations.

Contribution

It provides a comprehensive comparison of state-of-the-art style transfer models specifically for comic styles, including real-life validation via user surveys.

Findings

Different models have distinct advantages and disadvantages in comic style transfer.

Quantitative and qualitative analyses reveal trade-offs in style transfer quality.

User surveys support the evaluation results in practical applications.

Abstract

The work by Gatys et al. [1] recently showed a neural style algorithm that can produce an image in the style of another image. Some further works introduced various improvements regarding generalization, quality and efficiency, but each of them was mostly focused on styles such as paintings, abstract images or photo-realistic style. In this paper, we present a comparison of how state-of-the-art style transfer methods cope with transferring various comic styles on different images. We select different combinations of Adaptive Instance Normalization [11] and Universal Style Transfer [16] models and confront them to find their advantages and disadvantages in terms of qualitative and quantitative analysis. Finally, we present the results of a survey conducted on over 100 people that aims at validating the evaluation results in a real-life application of comic style transfer.