A Multi-domain Style Transfer by Modified Generator of GAN (GAN의 수정된 생성자에 의한 다중 도메인 스타일 트랜스퍼)

Geum-Boon Lee (이금분)

doi:10.9708/jksci.2022.27.07.027

A Multi-domain Style Transfer by Modified Generator of GAN

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2022, 27(7), pp.27-33
DOI : 10.9708/jksci.2022.27.07.027
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : May 12, 2022
Accepted : July 8, 2022
Published : July 29, 2022

Geum-Boon Lee ¹

¹조선대학교

Accredited

ABSTRACT

In this paper, we propose a novel generator architecture for multi-domain style transfer method not an image to image translation, as a method of generating a styled image by transfering a style to the content image. A latent vector and Gaussian noises are added to the generator of GAN so that a high quality image is generated while considering the characteristics of various data distributions for each domain and preserving the features of the content data. With the generator architecture of the proposed GAN, networks are configured and presented so that the content image can learn the styles for each domain well, and it is applied to the domain composed of images of the four seasons to show the high resolution style transfer results.

KEYWORDS

Multi domain, Latent vector, Generator architecture, Gaussian noise, Style transfer

Citation status

* References for papers published after 2023 are currently being built.

[confproc] A. Almahairi / 2018 / Augmented cyclegan: Learning many-to-many mappings from unpaired data / Proceeding of the 35th International Conference on Machine : 195~204

[other] X. Huang / 2018 / Multimodal unsupervised image-to-image translation

[other] Y. Choi / 2020 / StarGAN v2: Diverse Image Synthesis for Multiple Domains

[other] T. Karras / 2019 / A Style-Based Generator architecture for Generative Adeversarial Networks

[other] S. Na / 2019 / MISO: Mutual Information Loss with Stochastic Style Representaions for Multimodal Image-to-Image Translation

[other] J-Y Zhu / 2017 / Toward Multimodal Image-to-Image Translation

[other] P. Isola / 2017 / Image-to-Image Translation with Conditional Adversarial Networks

[other] J-Y. Zhu / 2018 / Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

[other] Q. Mao / 2019 / Mode Seeking Generative Adversarial Networks for Divese Image Synthesis

[other] Y. Choi / 2018 / StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation

[other] D. P. Kingma / 2014 / Auto-Encoding Variaitonal Bayes

[other] A. Odena / 2017 / Conditional Image Synthesis With Auxiliary Classifier GANs

[other] I. J. Goodfellow / 2014 / Generative Adversarial Nets

[confproc] L. A. Gatys / 2016 / Image Style Transfer Using Convolutional Neural Networks / Proceeding of IEEE Conference on Computer Vision and Pattern Recognition : 2414~2423

[other] E. Risser / 2017 / Stable and Controllable Neural Texture Synthesis and Style Transfer Using Histogram Losses

[other] F. Luan / 2017 / Deep Photo Style Transfer

[other] X. Huang / 2017 / Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization

[other] S. Loffe / 2015 / Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

[other] A. Radford / 2016 / Unsupervised representation learning with deep convolutional generative adversarial network

[other] D. Ylyanov / 2016 / Texture networks: Feed-forward synthesis of textures and stylized images

[other] D. Ulyanov / 2017 / Improved texture networks: Maximizing quality and diversity in feed-forward stylization and texture synthsesis

[other] Y. Shen / 2020 / InterFaceGAN:Interpreting the Disentangled Face Representation Learned by GANs

[confproc] J. Chang / 2019 / Layered Conceptual Image Compression Via Deep Semantic Synthesis / 2019. In Proceeding of the IEEE Conference on Image Processing : 694~698

[other] M. Heusel / 2018 / GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

This paper was written with support from the National Research Foundation of Korea.

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

A Multi-domain Style Transfer by Modified Generator of GAN

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

A Multi-domain Style Transfer by Modified Generator of GAN

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (24) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.