Light field super-resolution using complementary-view feature attention

作     者：Wei Zhang Wei Ke Da Yang Hao Sheng Zhang Xiong Wei Zhang;Wei Ke;Da Yang;Hao Sheng;Zhang Xiong

作者机构：Faculty of Applied SciencesMacao Polytechnic UniversityMacao SAR 999078China State Key Laboratory of Software Development EnvironmentSchool of Computer Science and EngineeringBeihang UniversityBeijing 100191Chinaand Beihang Hangzhou Innovation Institute YuhangXixi Octagon CityYuhang DistrictHangzhou 310023China 

出 版 物：《Computational Visual Media》 (计算可视媒体（英文版）)

年 卷 期：2023年第9卷第4期

页      面：843-858页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 0835[工学-软件工程] 

基　　金：supported by the National Key R&D Program of China(2018YFB2100500) the National Natural Science Foundation of China(61872025) the Science and Technology Development Fund,Macao SAR(0001/2018/AFJ) the Open Fund of the State Key Laboratory of Software Development Environment(SKLSDE-2021ZX-03). 

主　　题：light field(LF) super-resolution(SR) attention 

摘      要：Light field(LF)cameras record multiple perspectives by a sparse sampling of real scenes,and these perspectives provide complementary information.This information is beneficial to LF super-resolution(LFSR).Compared with traditional single-image super-resolution,LF can exploit parallax structure and perspective correlation among different LF views.Furthermore,the performance of existing methods are limited as they fail to deeply explore the complementary information across LF views.In this paper,we propose a novel network,called the light field complementary-view feature attention network(LF-CFANet),to improve LFSR by dynamically learning the complementary information in LF views.Specifically,we design a residual complementary-view spatial and channel attention module(RCSCAM)to effectively interact with complementary information between complementary views.Moreover,RCSCAM captures the relationships between different channels,and it is able to generate informative features for reconstructing LF images while ignoring redundant information.Then,a maximum-difference information supplementary branch(MDISB)is used to supplement information from the maximum-difference angular positions based on the geometric structure of LF images.This branch also can guide the process of reconstruction.Experimental results on both synthetic and real-world datasets demonstrate the superiority of our method.The proposed LF-CFANet has a more advanced reconstruction performance that displays faithful details with higher SR accuracy than state-of-the-art methods.