Multiform Fonts-to-Fonts Translation via Style and Content Disentangled Representations of Chinese Character
This addresses font generation for designers or users needing custom Chinese fonts, but it is incremental as it builds on existing style transfer methods.
The paper tackled generating personalized fonts by treating it as an image style transfer problem, designing a network to disentangle and recombine content and style from Chinese characters, enabling synthesis of full font sets with few examples; results showed generated characters closely matched real ones using SSIM and PSNR metrics.
This paper mainly discusses the generation of personalized fonts as the problem of image style transfer. The main purpose of this paper is to design a network framework that can extract and recombine the content and style of the characters. These attempts can be used to synthesize the entire set of fonts with only a small amount of characters. The paper combines various depth networks such as Convolutional Neural Network, Multi-layer Perceptron and Residual Network to find the optimal model to extract the features of the fonts character. The result shows that those characters we have generated is very close to real characters, using Structural Similarity index and Peak Signal-to-Noise Ratio evaluation criterions.