Abstract:Single image super-resolution (SISR), as a traditional ill-conditioned inverse problem, has been greatly revitalized by the recent development of convolutional neural networks (CNN). These CNN-based methods generally map a low-resolution image to its corresponding high-resolution version with sophisticated network structures and loss functions, showing impressive performances. This paper provides a new insight on conventional SISR algorithm, and proposes a substantially different approach relying on the iterative optimization. A novel iterative super-resolution network (ISRN) is proposed on top of the iterative optimization. We first analyze the observation model of image SR problem, inspiring a feasible solution by mimicking and fusing each iteration in a more general and efficient manner. Considering the drawbacks of batch normalization, we propose a feature normalization (F-Norm, FN) method to regulate the features in network. Furthermore, a novel block with FN is developed to improve the network representation, termed as FNB. Residual-in-residual structure is proposed to form a very deep network, which groups FNBs with a long skip connection for better information delivery and stabling the training phase. Extensive experimental results on testing benchmarks with bicubic (BI) degradation show our ISRN can not only recover more structural information, but also achieve competitive or better PSNR/SSIM results with much fewer parameters compared to other works. Besides BI, we simulate the real-world degradation with blur-downscale (BD) and downscale-noise (DN). ISRN and its extension ISRN+ both achieve better performance than others with BD and DN degradation models.

Image supper resolution by dilated dense progressive network

Single image resolution enhancement by efficient dilated densely connected residual network

Detail-preserving Image Super-Resolution Via Recursively Dilated Residual Network

Hyperspectral Image Super-Resolution Via Convolutional Neural Network

Image Super-Resolution Based on Dense Convolutional Network.

Fast Single Image Super-Resolution Via Dilated Residual Networks

Towards realistic image via function learning

End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks

Learning Depth Super-Resolution by Using Multi-Scale Convolutional Neural Network.

Image Superresolution Via Dense Discriminative Network.

Image super-resolution via dynamic network

Image Super-Resolution via Simplified Dense Network With Non-Degenerate Layers.

Image Super-Resolution Using Deep Convolutional Networks

Image Superresolution using Scale-Recurrent Dense Network

An Image Super-Resolution Deep Learning Network Based on Multi-Level Feature Extraction Module.

The super-resolution reconstruction algorithm of multi-scale dilated convolution residual network

Single image super-resolution via deep progressive multi-scale fusion networks

Super-Resolution for Noisy Images Via Deep Convolutional Neural Network

Image Super-Resolution via Deep Feature Recalibration Network.

Image Superresolution Using Densely Connected Residual Networks

Iterative Network for Image Super-Resolution