Sorry, you need to enable JavaScript to visit this website.

SINGLE IMAGE SUPER-RESOLUTION VIA GLOBAL-CONTEXT ATTENTION NETWORKS

Primary tabs

Citation Author(s):
Pengcheng Bian, Zhonglong Zheng, Dawei Zhang, Liyuan Chen, Minglu Li
Submitted by:
Pengcheng Bian
Last updated:
25 September 2021 - 2:12am
Document Type:
Poster
Document Year:
2021
Event:
Presenters Name:
Pengcheng Bian
Paper Code:
1950

Abstract 

Abstract: 

In the last few years, single image super-resolution (SISR) has benefited a lot from the rapid development of deep convolutional neural networks (CNNs), and the introduction of attention mechanisms further improves the performance of SISR. However, previous methods use one or more types of attention independently in multiple stages and ignore the correlations between different layers in the network. To address these issues, we propose a novel end-to-end architecture named global-context attention network (GCAN) for SISR, which consists of several residual global-context attention blocks (RGCABs) and an inter-group fusion module (IGFM). Specifically, the proposed RGCAB extracts representative features that capture non-local spatial interdependencies and multiple channel relations. Then the IGFM aggregates and fuses hierarchical features of multi-layers discriminatively by considering correlations among layers. Extensive experimental results demonstrate that our method achieves superior results against other state-of-the-art methods on publicly available datasets.

up
0 users have voted:

Dataset Files

Poster

(29)

Presentation

(31)