An end-to-end model for chinese calligraphy generation
- PDF / 1,734,099 Bytes
- 18 Pages / 439.642 x 666.49 pts Page_size
- 79 Downloads / 214 Views
An end-to-end model for chinese calligraphy generation Peichi Zhou1 · Zipeng Zhao1 · Kang Zhang2 · Chen Li1 · Changbo Wang1 Received: 2 July 2019 / Revised: 23 July 2020 / Accepted: 25 August 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract A Chinese calligraphy copybook usually has a limited number of Chinese characters, far from a whole set of characters needed for typesetting. Therefore, there is a need to develop complete sets of Chinese calligraphy libraries for well-known calligrapher styles. This paper proposes an end-to-end network for character generation based on specific calligraphy styles. Specifically, a style transfer network is designed to transfer the style of characters, and a content supplement network is designed to capture the details of stylish strokes. Our model can generate high-quality calligraphy images without manually annotating data. To verify the generated calligraphy styles, a new dataset is constructed for experimental comparison between our method and two other baseline methods. Moreover, a user study is conducted to evaluate our generated calligraphy from a visual perspective. When the experiment participants are asked to distinguish the real calligraphy from generated samples, the correct rate was 53.5%. The results show that the calligraphy styles generated by our model are almost indistinguishable from the original works. Keywords Calligraphy · Generative models · Font style transfer · Deep learning
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11042-020-09709-5) contains supplementary material, which is available to authorized users. Changbo Wang
[email protected] Peichi Zhou [email protected] Kang Zhang [email protected] 1
East China Normal University, Shanghai, China
2
University of Texas at Dallas, Richardson, TX, USA
Multimedia Tools and Applications
1 Introduction In ancient China, Chinese characters gradually evolved into an art form called calligraphy. Unlike painting and music, calligraphy is both the art of a given form and a common medium of communication. Calligraphy is widely used in graphic design, stone inscriptions, and product design. Ideally, individual calligraphy characters could be scanned into a font library, making it easy for calligraphy enthusiasts to learn and for designers to use. It is, however, impossible in reality, as none of the original calligraphy works contains the entire set of commonly used characters, minimally over 3000. The current solution involves designers mimicking the style of the calligraphy in manually making the missing characters. This manual process is tremendously time-consuming, since each Chinese character is constructed by one or more radicals. There are several ways to combine these radicals into a character, such as left-right structure, up-down structure, as shown in Fig. 1. The position of a radical also influences the style. The Chinese language has more than 80,000 logographic characters, in contrast to a phon
Data Loading...
