Principles of Automatic Generation of Artistic Chinese Calligraphy

Abstract

In this chapter we will introduce a novel intelligent system which can automatically generate new Chinese calligraphic artwork to meet visually aesthetic requirements. The system first extracts the hierarchical parametric representations of Chinese characters from input images of existing calligraphic style to form a compact set of training examples. Using a six-layer hierarchical representation, the extraction results are stored in a small structural stroke database, which are then exploited to form a continuous calligraphy knowledge space. The space is spanned by character examples of different styles (knowledge sources) which are aggregated and aligned according to a proposed constraint-based analogous reasoning process. By also incorporating a set of simple and yet effective geometric constraints, the proposed system can generate novel calligraphic styles that are aesthetically appealing. Samples of novel calligraphic artwork produced using the system are presented to demonstrate the effectiveness of our approach. The combination of knowledge from various input sources creates a huge space for the intelligent system to explore and produce new styles of calligraphy. Possible applications of the proposed system are also discussed. KeywordsChinese CharacterAutomatic GenerationParametric RepresentationAnalogous ReasoningHandwriting RecognitionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.