A WordArt Editing Method and System Combined with Stylization and Destylization

Abstract

The invention relates to a combined stylized and destylized artistic word editing method and system, belonging to the technical field of image texture synthesis. The steps of the method include: constructing a font encoder, a font decoder, a font effect encoder, a font effect decoder, a font discriminator and a font discriminator to form a joint network structure. In the training phase, the network is trained in three aspects: glyph encoding and decoding, art word image destylization and text image stylization. In the test phase, the network can complete three tasks: destylization, that is, input the wordart picture, and get the corresponding text picture; stylization, that is, input the reference wordart picture and the target text picture, and get the corresponding wordart picture; style creation : Use the font effect encoder to encode various reference font effects, perform weighted average of the encoding results, and obtain brand new font effects after decoding. The present invention can generate high-quality artistic word pictures by jointly considering font encoding and decoding, stylization and destylization.

Description

Technical field[0001]The invention belongs to the technical field of image texture synthesis, and relates to a method and system for editing artistic characters that combine stylization and de-stylization.Background technique[0002]Art characters are widely used in graphic design such as posters, magazine covers, advertising designs, etc. However, the manual design and editing process of art characters is tedious and requires certain skills. The fully automatic word art editing method has great commercial value.[0003]The key technology of word art synthesis is texture synthesis, which adds texture information representing the style of the word effect to the target glyph. Although texture synthesis is widely used, it lacks a suitable method for artistic word generation. The texture synthesis method based on statistics, which describes the texture as a global statistic, is not suitable for synthesizing the texture of WordArt with distinctive structure; while the texture synthesis metho...

AI Technical Summary

Problems solved by technology

Despite the widespread use of texture synthesis, methods suitable for WordArt generation are lacking

The texture synthesis method based on statistics, which describes the texture as a global statistic, is not suitable for synthesizing artistic text textures with distinct structures; while the texture synthesis method based on local blocks cannot guarantee the distribution of global textures

Images