Unified Style Transfer

TL;DR

The paper introduces the Unified Style Transfer (UST) model that unifies different style transfer approaches and proposes a new evaluation method based on style distribution consistency, addressing evaluation challenges in style transfer research.

Contribution

It presents a novel generative model for internal style representation and a new evaluation philosophy called Statistical Style Analysis, unifying style transfer methods and improving validation.

Findings

UST enables simultaneous domain-based and image-based style transfer.

Statistical Style Analysis offers a new way to validate style transfer models.

Discussion on translation-invariance of AdaIN features enhances understanding of style features.

Abstract

Currently, it is hard to compare and evaluate different style transfer algorithms due to chaotic definitions of style and the absence of agreed objective validation methods in the study of style transfer. In this paper, a novel approach, the Unified Style Transfer (UST) model, is proposed. With the introduction of a generative model for internal style representation, UST can transfer images in two approaches, i.e., Domain-based and Image-based, simultaneously. At the same time, a new philosophy based on the human sense of art and style distributions for evaluating the transfer model is presented and demonstrated, called Statistical Style Analysis. It provides a new path to validate style transfer models' feasibility by validating the general consistency between internal style representation and art facts. Besides, the translation-invariance of AdaIN features is also discussed.