Skip to main navigation Skip to search Skip to main content

Chinese Calligraphy Style Transfer with Generative Adversarial Network

  • Jiani Dai*
  • *Corresponding author for this work
  • University of Chinese Academy of Sciences
  • Nanyang Technological University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This manuscript delineates an innovative approach for automating the emulation and creation of Chinese calligraphy styles using Generative Adversarial Networks (GANs). The proposed framework CCUT consists of two main elements: A content supplementation network and a style transfer network. These components work collaboratively to learn from two of Chinese calligraphy masters, Zhao Mengfu and Yan Zhenqin, and transform standard Zheng Kai font into a unique calligraphy style while retaining the essential textual content. The objective of this research is to reduce manual labor involved in the transfer of calligraphic styles and to investigate its prospective applications in the development of fonts. Additionally, we conduct a comprehensive evaluation of the outcomes generated by our and others' models in comparison to human calligraphy, offering an in-depth analysis of the discrepancies noted.

Original languageEnglish
Title of host publicationARAEML 2024 - 2024 International Conference on Advanced Robotics, Automation Engineering and Machine Learning
Subtitle of host publicationConference Proceeding
PublisherAssociation for Computing Machinery
Pages104-108
Number of pages5
ISBN (Electronic)9798400717116
DOIs
StatePublished - 28 Jun 2024
Externally publishedYes
Event2024 International Conference on Advanced Robotics, Automation Engineering and Machine Learning, ARAEML 2024 - Hangzhou, China
Duration: 28 Jun 202430 Jun 2024

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2024 International Conference on Advanced Robotics, Automation Engineering and Machine Learning, ARAEML 2024
Country/TerritoryChina
CityHangzhou
Period28/06/2430/06/24

Keywords

  • Calligraphy
  • Deep Learning
  • Generative Adversarial Networks
  • Image-To-image Translation
  • Style Transfer

Fingerprint

Dive into the research topics of 'Chinese Calligraphy Style Transfer with Generative Adversarial Network'. Together they form a unique fingerprint.

Cite this