回到写日记的年月里，fight！

本文是对基于深度学习的图像超分辨率的一篇近期回顾性论文的详细研读。

In this document we will perform two types of transfer learning: finetuning and feature extraction.

本文提出了一种新型的Simple Image Super Resolution(SISR)框架，在前人将生成对抗网络(GANs)应用到SISR领域的基础上进行了结构创新，经过实验证明了该框架达到了最前沿的(state-of-the-art)研究结果，产生了非常好的图像恢复效果。

What text editor to use? Sass or plain old CSS? What on earth is Compass? Command line? I’m not touching that. Sound like you? Welcome, I was once like you and this is the guide I wish someone had given me.