Recently, with the revolutionary neural style transferring methods,creditable paintings can be synthesized automatically from content images andstyle images. However, when it comes to the task of applying a painting's styleto an anime sketch, these methods will just randomly colorize sketch lines asoutputs and fail in the main task: specific style tranfer. In this paper, weintegrated residual U-net to apply the style to the gray-scale sketch withauxiliary classifier generative adversarial network (AC-GAN). The whole processis automatic and fast, and the results are creditable in the quality of artstyle as well as colorization.
展开▼