Open Access iconOpen Access

ARTICLE

Explicitly Color-Inspired Neural Style Transfer Using Patchified AdaIN

Bumsoo Kim1, Wonseop Shin2, Yonghoon Jung1, Youngsup Park3, Sanghyun Seo1,4,*

1 Department of Applied Art and Technology, Chung-Ang University, Anseong, 17546, Republic of Korea
2 Department of Advanced Imaging Science Multimedia & Film, Chung-Ang University, Seoul, 06974, Republic of Korea
3 Innosimulation Co., Ltd., Gangseo-gu, 07794, Republic of Korea
4 School of Art and Technology, Chung-Ang University, Anseong, 17546, Republic of Korea

* Corresponding Author: Sanghyun Seo. Email: email

Computer Modeling in Engineering & Sciences 2024, 141(3), 2143-2164. https://doi.org/10.32604/cmes.2024.056079

Abstract

Arbitrary style transfer aims to perceptually reflect the style of a reference image in artistic creations with visual aesthetics. Traditional style transfer models, particularly those using adaptive instance normalization (AdaIN) layer, rely on global statistics, which often fail to capture the spatially local color distribution, leading to outputs that lack variation despite geometric transformations. To address this, we introduce Patchified AdaIN, a color-inspired style transfer method that applies AdaIN to localized patches, utilizing local statistics to capture the spatial color distribution of the reference image. This approach enables enhanced color awareness in style transfer, adapting dynamically to geometric transformations by leveraging local image statistics. Since Patchified AdaIN builds on AdaIN, it integrates seamlessly into existing frameworks without the need for additional training, allowing users to control the output quality through adjustable blending parameters. Our comprehensive experiments demonstrate that Patchified AdaIN can reflect geometric transformations (e.g., translation, rotation, flipping) of images for style transfer, thereby achieving superior results compared to state-of-the-art methods. Additional experiments show the compatibility of Patchified AdaIN for integration into existing networks to enable spatial color-aware arbitrary style transfer by replacing the conventional AdaIN layer with the Patchified AdaIN layer.

Keywords


Cite This Article

APA Style
Kim, B., Shin, W., Jung, Y., Park, Y., Seo, S. (2024). Explicitly color-inspired neural style transfer using patchified adain. Computer Modeling in Engineering & Sciences, 141(3), 2143-2164. https://doi.org/10.32604/cmes.2024.056079
Vancouver Style
Kim B, Shin W, Jung Y, Park Y, Seo S. Explicitly color-inspired neural style transfer using patchified adain. Comput Model Eng Sci. 2024;141(3):2143-2164 https://doi.org/10.32604/cmes.2024.056079
IEEE Style
B. Kim, W. Shin, Y. Jung, Y. Park, and S. Seo, “Explicitly Color-Inspired Neural Style Transfer Using Patchified AdaIN,” Comput. Model. Eng. Sci., vol. 141, no. 3, pp. 2143-2164, 2024. https://doi.org/10.32604/cmes.2024.056079



cc Copyright © 2024 The Author(s). Published by Tech Science Press.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 306

    View

  • 72

    Download

  • 0

    Like

Related articles

Share Link