Chinese Font Translation with Improved Wasserstein Generative Adversarial Network

摘要

Nowadays, various fonts are applied in many fields, and the generation of multiple fonts by computer plays animportant role in the inheritance, development and innovation of Chinese culture. Aiming at the existing font generationmethods, which have some problems such as stroke deletion, artifact and blur, this paper proposes Chinese fonttranslation with improved wasserstein generative adversarial network. The wasserstein distance is used to measuredistance and difference between the two distributions, and the gradient penalty mechanism is used instead of the weightclipping. The residual dense blocks with better flexibility are selected as the core component to extract the features fullyand enhance the information transmission between network layers. It realizes the style migration between differentChinese fonts. Experiments show that the proposed method has better performance in font generation details, simplifiesthe font translation process, and improves the fidelity of font generation.