A fast solution for Chinese calligraphy relief modeling from 2D handwriting image
- PDF / 3,852,515 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 41 Downloads / 176 Views
ORIGINAL ARTICLE
A fast solution for Chinese calligraphy relief modeling from 2D handwriting image Yu-Wei Zhang1 · Jinlei Wang1 · Wenfei Long1 · Hui Liu2 · Caiming Zhang3 · Yanzhao Chen1
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Calligraphy occupies a distinguished position in Chinese traditional culture. For long-term preservation, many calligraphy works have been carved on stones or woods in the form of relief. In this paper, we present a novel solution that enables fast modeling of Chinese calligraphy relief from 2D handwriting image, which benefits from the advances of deep learning. We first construct a relief dataset composed of diverse types of calligraphy fonts and then design a convolutional neural network for height predictions. Through the trained network, one can quickly generate homogeneous type, inhomogeneous type or hybrid style of reliefs. The advantage over previous methods is that our method does not require parameter tuning and is fast in generating calligraphy reliefs from different resolution of inputs. A number of experiments and comparisons prove the effectiveness of our method. Keywords Chinese calligraphy · Relief modeling · 3D font
1 Introduction Calligraphy, or the art of writing, is one of the soul forms in Chinese culture. It is usually written on paper by using ink brushes and conveys the feeling of beauty and personality.
B
Yanzhao Chen [email protected] Yu-Wei Zhang [email protected] Jinlei Wang [email protected] Wenfei Long [email protected] Hui Liu [email protected] Caiming Zhang [email protected]
1
School of Mechanical and Automotive Engineering, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China
2
School of Computer Science and Technology, Shandong University of Finance and Economics, Jinan, China
3
School of Computer Science and Technology, Shandong University, Jinan, China
For long-term preservation, many calligraphy works have been carved on stones or woods in the form of relief. The key problem of calligraphy relief modeling is how to inflate the font regions in 3D manner. One simple way is to extrude the fonts to a predefined height, but the result is limited to be flat. To produce stronger sense of 3Dness, some software packages such as ArtForm and JDPaint have provided specific tools to generate calligraphy relief in continuous and smooth form. However, heavy manual operations are required when using these tools. Before our work, several automatic methods have been proposed to produce calligraphy reliefs. One of them [1] lifts the fonts first and then applies Gaussian blurring to smooth the stroke edges. Through multiple iterations, the relief surface becomes smooth and bumpy. This method produces acceptable result, but is time-consuming and some parameters need to be carefully tuned. To avoid parameter tuning, one recent work [2] proposes to model calligraphy reliefs though Laplacian-based height inflation and normalbased height inflation. However, the reconstruction relies on solving several spa
Data Loading...
