Lightweight Image Super-Resolution with Enhanced CNN(LESRCNN)is conducted by Chunwei Tian, Ruibin Zhuge, Zhihao Wu, Yong Xu, Wangmeng Zuo, Chen Chen and Chia-Wen Lin, and accepted by Knowledge-Based Systems in 2020. It is implemented by Pytorch. And it is reported by Cver and 52CV. Its website is https://mp.weixin.qq.com/s/njlAEQXxjXKqFcxM7KYiqA.
This paper uses a flexible sub-pixel convolutional technique for image blind super-resolution, which is useful for phones and cameras. Also, it has less parameters and faster super-resolution speed.
Abstract
Deep convolutional neural networks (CNNs) with strong expressive ability have achieved impressive performances on single image super-resolution (SISR). However, their excessive amounts ofconvolutions and parameters usually consume high computational cost and more memory storagefor training a SR model, which limits their applications to SR with resource-constrained devicesin real world. To resolve these problems, we propose a lightweight enhanced SR CNN (LESRCNN) with three successive sub-blocks, an information extraction and enhancement block (IEEB), a reconstruction block (RB) and an information refinement block (IRB). Specifically, the IEEB extracts hierarchical low-resolution (LR) features and aggregates the obtained features step-by-step to increase the memory ability of the shallow layers on deep layers for SISR. To remove redundant information obtained, a heterogeneous architecture is adopted in the IEEB. After that, the RB converts low-frequency features into high-frequency features by fusing global and local features, which is complementary with the IEEB in tackling the long-term dependency problem. Finally,the IRB uses coarse high-frequency features from the RB to learn more accurate SR features and construct a SR image. The proposed LESRCNN can obtain a high-quality image by a model fordifferent scales. Extensive experiments demonstrate that the proposed LESRCNN outperforms state-of-the-arts on SISR in terms of qualitative and quantitative evaluation.