Context Reasoning Attention Network for Image Super-Resolution

Deep convolutional neural networks (CNNs) are achieving great successes for image super-resolution (SR), where global context is crucial for accurate restoration. However, the basic convolutional layer in CNNs is designed to extract local patterns, lacking the ability to model global context. With global context information, lots of efforts have been devoted to augmenting SR networks, especially by global feature interaction methods. These works incorporate the global context into local feature representation. However, recent advances in neuroscience show that it is necessary for the neurons to dynamically modulate their functions according to context, which is neglected in most CNN based SR methods. Motivated by those observations and analyses, we propose context reasoning attention network (CRAN) to modulate the convolution kernel according to the global context adaptively. Specifically, we extract global context descriptors, which are further enhanced with semantic reasoning. Channel and spatial interactions are then introduced to generate context reasoning attention mask, which is applied to modify the convolution kernel adaptively. Such a modulated convolution layer is utilized as basic component to build the blocks and networks. Extensive experiments on benchmark datasets with multiple degradation models show that CRAN obtains superior results and favorable trade-off between performance and model complexity.

1. Introduction

Image super-resolution (SR) aims to reconstruct an accurate high-resolution (HR) image given its low-resolution (LR) counterpart [14]. Image SR plays a fundamental role in various computer vision applications, ranging from security and surveillance imaging [71], medical imaging [48], to object recognition [45]. However, image SR is an ill-posed problem, since there exists multiple solutions for any LR input. To tackle such an inverse problem, lots of deep convolutional neural networks (CNNs) have been proposed to learn mappings between LR and HR image pairs.