介绍(由于近期准备校招,博客暂时不更新)
神经风格迁移是我研一研二时期主要的研究方向,而从最初的风格迁移出现已经有较长一段时间了。之所以现在写这个博客,第一是因为我的毕业论文定的方向是风格迁移+情感分析;第二是借这篇博客以及之后的学习,能对深度学习更进一步的理解!
目录
什么是风格迁移?
-
- α 图定义为
style image
, p 图定义为content image
,
- α 图定义为
-
- 损失通过VGG-16的前四层来表示,层次越高,内容越抽象。这里列出几个符号定义。
-
- 将内容图像输入卷积网络中提取图像内容,由公式
,计算内容损失。
-
- 对以上公式求导
,使用反向传输,使得生成图像在内容
上接近原输入内容图像。
-
- 将风格图像输入到同一个网络中提取它的风格信息,风格提取的符号定义为
-
- 计算风格图像的loss
单独某层的损失函数
各层综合的损失函数
求偏导
,使得生成图像在风格
上接近原输入风格图像。
- 7.风格损失
- 8.
Gatys-Image-Style-Transfer
中给出的流程图。X是白噪声图像。同时将三张图片输入到同一网络中,对内容图像和风格图像求特征,对白噪声X求导。
当前不同框架下的风格迁移
几年前,Gatys等人的风格迁移
[paper],[code]在学术界引起了不错的反响,并催生了后续很多研究成果。虽然在Gatys之前已经有学者做迁移方面的研究,但我把这篇paper看作是first style transfer paper
。
1.基于图像优化的Slow Transfer
- [A Neural Algorithm of Artistic Style ][paper]
- [Image Style Transfer Using Convolutional Neural Networks] [Paper] (CVPR 2016)
-
[Demystifying Neural Style Transfer][paper](Theoretical Explanation) (IJCAI 2017)
-
[Stable and Controllable Neural Texture Synthesis and Style Transfer Using Histogram Losses][paper]
- [Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis][paper](CVPR 2016)
2.基于模型优化的Fast Transfer
2.1Per-Style-Per-Model-Methods
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution][paper] (ECCV 2016)
- [Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks] [Paper] (ECCV 2016)
-
[Texture Networks: Feed-forward Synthesis of Textures and Stylized Images] [Paper] (ICML 2016)
- [Improved Texture Networks: Maximizing Quality and Diversity in Feed-forward Stylization and Texture Synthesis] [Paper] (CVPR 2017)
- [Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks] [Paper] (ECCV 2016)
- Torch-based
2.2Multiple-Style-Per-Model-Methods
-
[A Learned Representation for Artistic Style] [Paper] (ICLR 2017)
-
[Multi-style Generative Network for Real-time Transfer] [Paper]
-
[StyleBank: An Explicit Representation for Neural Image Style Transfer] [Paper] (CVPR 2017)
-
[Diversified Texture Synthesis With Feed-Forward Networks] [Paper] (CVPR 2017)
2.3Arbitrary-Style-Per-Model-Methods
- [Fast Patch-based Style Transfer of Arbitrary Style] [Paper]
-
[Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization] [Paper] (ICCV 2017)