摘要
为推进基于深度学习的图像风格迁移的技术研究,对目前基于深度学习的图像风格迁移的主要方法和代表性工作进行了归纳与探讨。回顾了非参数的图像风格迁移,详细介绍了目前主要的基于深度学习的图像风格迁移的基本原理和方法,分析了图像风格迁移在相关领域中的应用前景,最后总结了基于深度学习的图像风格迁移目前存在的问题与未来的研究方向。
In order to promote the technology research of image style transfer based on deep learning,this paper summarized and discussed the current major methods and representative work. Firstly,this paper reviewed the non-parametric image style transfer,and introduced the basic principles and methods of image style transfer based on deep learning,and analyzed the application prospect of image style transfer technology in related fields. At last,this paper summarized the existing problems and future research directions of image style transfer based on deep learning.
作者
陈淑環
韦玉科
徐乐
董晓华
温坤哲
Chen Shuhuan;Wei Yuke;Xu Le;Dong Xiaohua;Wen Kunzhe(School of Computer Science,Guangdong University of Technology,Guangzhou 510006,China)
出处
《计算机应用研究》
CSCD
北大核心
2019年第8期2250-2255,共6页
Application Research of Computers
基金
广东省省级科技计划资助项目(2014A040402007)
关键词
图像风格迁移
深度学习
迁移学习
纹理合成
image style transfer
deep learning
transfer learning
texture synthesis
