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.

Benchmark of deep learning models for single image super-resolution (SISR)

Epistemic-Uncertainty-Based Divide-and-Conquer Network for Single-Image Super-Resolution

A Deep Learning Based No-Reference Image Quality Assessment Model For Single-Image Super-Resolution

Network architecture for single image super‐resolution: A comprehensive review and comparison

A Systematic Survey of Deep Learning-based Single-Image Super-Resolution

Orientation-Aware Deep Neural Network for Real Image Super-Resolution.

Deep Learning for Single Image Super-Resolution: A Brief Review

Fast and Accurate Single Image Super-Resolution Via an Energy-Aware Improved Deep Residual Network.

Deep Objective Quality Assessment Driven Single Image Super-Resolution

HNSR: Highway Networks Based Deep Convolutional Neural Networks Model for Single Image Super-Resolution

A review of single image super-resolution reconstruction based on deep learning

Single-image Super-Resolution Via Selective Multi-Scale Network

Efficient Deep Models for Real-Time 4K Image Super-Resolution. NTIRE 2023 Benchmark and Report.

ESKN: Enhanced Selective Kernel Network for Single Image Super-Resolution

An Attention-Based Approach for Single Image Super Resolution.

Single Image Super-Resolution Methods: A Survey

A Deep Journey into Super-resolution: A survey

Iterative Network for Image Super-Resolution

Real-world single image super-resolution: A brief review

Ultra Sharp : Study of Single Image Super Resolution using Residual Dense Network