Example-based Chinese Calligraphy Synthesis

DOI

10.2991/acaai-18.2018.59How to use a DOI?

Keywords

Chinese calligraphy synthesis; Chinese character topology; recurrent neural networks; optimization

Abstract

Chinese calligraphy synthesis is important meaningful to protect non-heritage cultural and augmented reality. In this paper, we proposed an algorithm of Chinese Calligraphy synthesis based on examples. First, we construct a stroke library represented by vector. Second, we obtain the trajectory of character stroke from the digital tablet input, and match the most similar character stroke from the stroke library and replace the trajectory with it. Third, we evaluate the style using recurrent neural networks (RNNs). We also propose a new model to describe Chinese character topology by 9-intersections & range relation (9IRR) according to experience. Finally, we employ simulated annealing strategy to optimize the Chinese character topology with aesthetic requirements. Experimental results show that character shapes generated are visual pleasing.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Wei Li
AU  - Yumin Chen
AU  - Chao Tang
AU  - Shaoyong Yu
PY  - 2018/03
DA  - 2018/03
TI  - Example-based Chinese Calligraphy Synthesis
BT  - Proceedings of the 2018 International Conference on Advanced Control, Automation and Artificial Intelligence (ACAAI 2018)
PB  - Atlantis Press
SP  - 253
EP  - 256
SN  - 1951-6851
UR  - https://doi.org/10.2991/acaai-18.2018.59
DO  - 10.2991/acaai-18.2018.59
ID  - Li2018/03
ER  -