SDANet: A stacked dense adaptive network for real-time and accurate super resolution

• Published in: Neurocomputing (Amsterdam) Vol. 370; pp. 70 - 77
• Main Author: Gao, Lizhao
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 22.12.2019
• Subjects: Adaptive learning; Convolutional neural network; Dense connection; Super resolution; Adaptive learning; Dense connection; Convolutional neural network; Super resolution
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2019.08.048

Recent advances in single-image super-resolution are mostly based on convolutional neural networks. However, the high computational cost is a critical obstacle to achieve real-time performance(24 fps). In this paper, we proposed the Stacked Dense Adaptive Network (SDANet) to reduce the computational cost and improve the computational density of CPUs/GPUs. Especially, SDANet consists of (1) a dense adaptive learning (DAL) module, which harnesses adaptive gate units (AGU) to retain the high-frequency information in the prior layers. (2) a stacked bottleneck (STB) module, that stacks the information of all previous layers to improve the computational density and reduce the computation dramatically. (3) a spatial-channel transform (SCT) module, which transforms bilinear spatial information to channels to improve the computational density via parallel computing. With the proposed three modules, our algorithm takes full advantage of the high-frequency information of all the layers, resulting in reducing the computational complexity dramatically. Extensive experiments verify the necessity of the SDANet, not only yielding the state-of-the-art results with a considerable margin on five benchmark datasets in terms of PSNR and visual quality but also achieving real-time performance( > 24 fps) with a generic CPU.