Lightweight Image Super-Resolution with Adaptive Weighted Learning Network

Abstract

Deep learning has been successfully applied to the single-image super-resolution (SISR) task with great performance in recent years. However, most convolutional neural network based SR models require heavy computation, which limit their real-world applications. In this work, a lightweight SR network, named Adaptive Weighted Super-Resolution Network (AWSRN), is proposed for SISR to address this issue. A novel local fusion block (LFB) is designed in AWSRN for efficient residual learning, which consists of stacked adaptive weighted residual units (AWRU) and a local residual fusion unit (LRFU). Moreover, an adaptive weighted multi-scale (AWMS) module is proposed to make full use of features in reconstruction layer. AWMS consists of several different scale convolutions, and the redundancy scale branch can be removed according to the contribution of adaptive weights in AWMS for lightweight network. The experimental results on the commonly used datasets show that the proposed lightweight AWSRN achieves superior performance on x2, x3, x4, and x8 scale factors to state-of-the-art methods with similar parameters and computational overhead. Code is avaliable at: https://github.com/ChaofWang/AWSRN

9 pages, 6 figures

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Sketch-based Manga Retrieval using Manga109 Dataset
• Wide Activation for Efficient and Accurate Image Super-Resolution
• Deep Learning for Image Super-resolution: A Survey
• Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search
• Residual Convolutional Neural Network Revisited with Active Weighted Mapping

Cited by in corpus (11)

• Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search
• Hybrid Residual Attention Network for Single Image Super Resolution
• Attention in Attention Network for Image Super-Resolution
• A Systematic Survey of Deep Learning-based Single-Image Super-Resolution
• Lightweight Single-Image Super-Resolution Network with Attentive Auxiliary Feature Learning
• Multi-Attention Based Ultra Lightweight Image Super-Resolution
• Designing and Training of A Dual CNN for Image Denoising
• A High-Performance Accelerator for Super-Resolution Processing on Embedded GPU
• BAM: A Balanced Attention Mechanism for Single Image Super Resolution
• FC${}^2$N: Fully Channel-Concatenated Network for Single Image Super-Resolution
• Adaptive Densely Connected Super-Resolution Reconstruction