BriefGPT.xyz
May, 2022
通过对比学习增强领域任意图像风格转移
Domain Enhanced Arbitrary Image Style Transfer via Contrastive Learning
HTML
PDF
Yuxin Zhang, Fan Tang, Weiming Dong, Haibin Huang, Chongyang Ma...
TL;DR
本文提出一种使用深度对比学习的方法,通过学习图像特征来获得图像的样式表示,并使用生成网络进行样式转移。研究表明,相对于现有的基于二阶统计量的方法,这种方法能够更加有效地获取图像的样式信息并生成更好的结果。
Abstract
In this work, we tackle the challenging problem of
arbitrary image style transfer
using a novel
style feature representation
learning method. A suitable style representation, as a key component in image stylizati
→