Fusing bi-directional global–local features for single image super-resolution

被引：0

作者：

Hwang, Kyomin ^{[1
]}

Yoon, Gangjoon ^{[2
]}

Song, Jinjoo ^{[3
]}

Yoon, Sang Min ^{[3
]}

机构：

[1] Department of Intelligence and Information Science, Graduate School of Convergence Science and Technology, Seoul National University, 1 Gwanak-ro, Seoul,08826, Korea, Republic of

[2] National Institute for Mathematical Sciences, 70, Yuseong-daero 1689 beon-gil, Yuseong-gu, Daejeon,34047, Korea, Republic of

[3] HCI Lab., College of Computer Science, Kookmin University, 77 Jeongneung-ro, Seoul,02707, Korea, Republic of

来源：

Engineering Applications of Artificial Intelligence | 2024年 / 127卷

基金：

新加坡国家研究基金会;

关键词：

Image enhancement - Learning systems - Optical resolving power - Stairs - Textures;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Image super resolution, which obtains high resolution output from a corresponding low resolution image, has been challenging due to inefficiencies in establishing complex high dimensional mapping for massive raw data. Single image super resolution can dramatically improve performance compared with current algorithms due to the proliferation of deep learning systems. However, convolutional kernels in deep neural networks are locally connected to the input feature maps, whereas features only interact with their local neighbors. Mutual interference between local features without considering global features causes blurring and staircase effects. This paper proposes an end-to-end single image super resolution model by simultaneously separating high and low frequency features and learning adaptive local and global features to effectively reconstruct the high resolution image by minimizing the loss of edges and texture information. The image frequency decomposition module with an attention block emphasizes self-representative low frequency features to separate high and low frequency features. The bidirectional global and local feature exchange module extracts global and local features from the separated network and fuses each feature to improve performance. Quantitative and qualitative analyses for the proposed frequency adaptive network validated that the proposed method is stable and robust against blurring and staircase effects by separating texture and the structure into adaptive and shared networks. © 2023 Elsevier Ltd

引用