论文标题
使用凸优化技术的超级分辨率神经网络的自适应损失函数
Adaptive Loss Function for Super Resolution Neural Networks Using Convex Optimization Techniques
论文作者
论文摘要
单图像超分辨率（SISR）任务是指从低分辨率图像到相应的高分辨率图像学习映射。众所周知，这项任务非常困难，因为这是一个不适的问题。最近，卷积神经网络（CNN）在SISR上取得了最先进的表现。但是，CNN产生的图像不包含图像的细节。生成对抗网络（GAN）旨在解决此问题并恢复尖锐的细节。然而，众所周知，甘斯很难训练。除此之外，它们还会在高分辨率图像中产生伪影。在本文中，我们提出了一种方法，其中CNN试图在不同空间中而不是像素空间对齐图像。这样的空间是使用凸优化技术设计的。鼓励CNN学习图像的高频组件以及低频组件。我们已经表明，所提出的方法可以恢复图像的细节，并且在训练过程中是稳定的。
Single Image Super-Resolution (SISR) task refers to learn a mapping from low-resolution images to the corresponding high-resolution ones. This task is known to be extremely difficult since it is an ill-posed problem. Recently, Convolutional Neural Networks (CNNs) have achieved state of the art performance on SISR. However, the images produced by CNNs do not contain fine details of the images. Generative Adversarial Networks (GANs) aim to solve this issue and recover sharp details. Nevertheless, GANs are notoriously difficult to train. Besides that, they generate artifacts in the high-resolution images. In this paper, we have proposed a method in which CNNs try to align images in different spaces rather than only the pixel space. Such a space is designed using convex optimization techniques. CNNs are encouraged to learn high-frequency components of the images as well as low-frequency components. We have shown that the proposed method can recover fine details of the images and it is stable in the training process.
