【图像风格转换】项目参考资料总结

实现图像风格转换、神经风格图像的一些资料和实现。

1.原理篇

A.A Neural Algorithm of Artistic Style，风格迁移开山之作实现，对将固定风格迁移到固定内容上：

B.Perceptual Losses for Real-Time Style Transfer and Super-Resolution，预训练模型改进迁移速度，利用固定风格图像快速转换内容图像：

C.Meta Networks for Neural Style Transfer风格迁移原网络实现任意风格迁移网络的构建和快速风格化。

code:https://github.com/FalongShen/styletransfer

ref:
http://export.arxiv.org/abs/1811.08668（这篇文章的参考文献很好）
https://amds123.github.io/2018/11/21/Computational-Decomposition-of-Style-for-Controllable-and-Enhanced-Style-Transfer/

2.博客详解

风格迁移三部曲
StyleTransferTrilogy

ref:
各种融合项目总结https://blog.csdn.net/u014636245/article/details/85479698

3.参考项目总结

• Keras Implementation of Neural Style Transfer from the paper：Neural-Style-Transfer & Neural Doodles
  

• For the paperPerceptual Losses for Real-Time Style Transfer and Super-Resolution :fast-neural-style
  
  realtime :
  
  ref:https://blog.csdn.net/qq_40523737/article/details/83066107 from source: https://github.com/jcjohnson/fast-neural-style

• 《深度学习框架PyTorch：入门与实践》第八章的配套代码，实现了快速风格迁移
  

• fast-neural-style-tensorflow, A tensorflow implementation for Perceptual Losses for Real-Time Style Transfer and Super-Resolution.
  
  This code is based on Tensorflow-Slim and OlavHN/fast-neural-style.
  web端实现ref:https://blog.csdn.net/pirage/article/details/86685963

• 基于TensorFlow.js 的 Arbitrary style transfer
  

• Fritz Style Transfer，移动端的超小模型。
  

• opencv 3.4集成的快速风格迁移实现， 以及Neural Style Transfer with OpenCV ,来自书：Deep Learning for Computer Vision with Python.
  ref:https://blog.csdn.net/juebai123/article/details/86545556
  ref移动端：https://github.com/fritzlabs/fritz-models
  ref github search:https://github.com/search?o=desc&p=3&q=neural+style&s=stars&type=Repositories

---

pic from pexels.com