ImageStyle Transfer Using Convolutional Neural Networks
2016-10-26 22:23
555 查看
论文里可不光给了转换的样例例,还给了实现原理图:
左边这个是对大师的作品进行采样处理(学习),用了一个VGG-19的经典图形识别的卷积网络。
右边这个和左边极为类似,只不不过它是按步骤学习了了输入图片的⻛风格,也对应生成了了池化层,我们叫它照片矩阵吧。
中间这个网络是对各层的大师矩阵和照片矩阵进行比对,把这个比对的“差异”进行了量化,定义为损失函数Ltotal,显然Ltotal越大说明差距越大,Ltotal越小就说明差距越小。
左边这个是对大师的作品进行采样处理(学习),用了一个VGG-19的经典图形识别的卷积网络。
右边这个和左边极为类似,只不不过它是按步骤学习了了输入图片的⻛风格,也对应生成了了池化层,我们叫它照片矩阵吧。
中间这个网络是对各层的大师矩阵和照片矩阵进行比对,把这个比对的“差异”进行了量化,定义为损失函数Ltotal,显然Ltotal越大说明差距越大,Ltotal越小就说明差距越小。
相关文章推荐
- 风格迁移系列-Image Style Transfer Using Convolutional Neural Networks
- 论文总结 - Image Style Transfer Using Convolutional Neural Networks
- 《image Style Transfer Using Convolutional Neural Networks》论文笔记
- 图像风格迁移-Image Style Transfer Using Convolutional Neural Networks
- 卷积神经网络图像风格转移 Image StyleTransfer Using Convolutional Neural Networks
- Image Style Transfer Using Convolutional Neural Network(理论篇)
- Image Style Transfer Using Convolutional Neural Network
- Learning_and_Transferring_Mid-Level_Image_Representations_using_Convolutional_Neural_Networks
- 深度卷积神经网络用于图像缩放Image Scaling using Deep Convolutional Neural Networks
- learning and transferring mid-level image representations using convolutional neural networks
- Image Scaling using Deep Convolutional Neural Networks
- Learning and Transferring Mid-Level Image Representations using Convolutional Neural Networks
- 深度卷积神经网络用于图像缩放Image Scaling using Deep Convolutional Neural Networks
- XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks随手记
- 论文阅读:Learning and Transferring Mid-Level Image Representations using Convolutional Neural Networks
- XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
- CNN网络二值化--XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
- XNOR-Net ImageNet Classification Using Binary Convolutional Neural Networks
- 深度卷积神经网络用于图像缩放Image Scaling using Deep Convolutional Neural Networks
- Convolutional Neural Networks for No-Reference Image Quality Assessment 笔记