论文标题
真实图像超分辨率
Deep Cyclic Generative Adversarial Residual Convolutional Networks for Real Image Super-Resolution
论文作者
论文摘要
最近基于深度学习的单图像超分辨率（SISR）方法主要是在干净的数据域中训练其模型，在干净的数据域中，低分辨率（LR）和高分辨率（HR）图像来自由于双孔下采样假设而导致的无噪声设置（相同域）。但是，这种降解过程在现实世界中不可用。我们考虑一个深度的循环网络结构，以维持LR和HR数据分布之间的域一致性，该结构的灵感来自于Cyclegan在图像到图像翻译应用程序中的最新成功。我们通过以端到端的方式使用生成性对抗网络（GAN）框架训练，提出了超分辨率残差循环生成对抗网络（SRRESCYCGAN）。我们在定量和定性实验中演示了我们提出的方法，这些方法可以很好地推广到真实图像超分辨率，并且很容易为移动/嵌入式设备部署。此外，我们的SR结果在AIM 2020真实图像SR挑战数据集上表明，所提出的SR方法可与其他最先进的方法获得可比的结果。
Recent deep learning based single image super-resolution (SISR) methods mostly train their models in a clean data domain where the low-resolution (LR) and the high-resolution (HR) images come from noise-free settings (same domain) due to the bicubic down-sampling assumption. However, such degradation process is not available in real-world settings. We consider a deep cyclic network structure to maintain the domain consistency between the LR and HR data distributions, which is inspired by the recent success of CycleGAN in the image-to-image translation applications. We propose the Super-Resolution Residual Cyclic Generative Adversarial Network (SRResCycGAN) by training with a generative adversarial network (GAN) framework for the LR to HR domain translation in an end-to-end manner. We demonstrate our proposed approach in the quantitative and qualitative experiments that generalize well to the real image super-resolution and it is easy to deploy for the mobile/embedded devices. In addition, our SR results on the AIM 2020 Real Image SR Challenge datasets demonstrate that the proposed SR approach achieves comparable results as the other state-of-art methods.
