CASIA OpenIR  > 智能制造技术与系统研究中心  > 多维数据分析
Facial image super-resolution guided by adaptive geometric features
Fan, Zhenfeng1,2; Hu, Xiyuan1,2; Chen, Chen1; Wang, Xiaolian1; Peng, Silong1
Source PublicationEURASIP Journal on Wireless Communications and Networking
ISSN1687-1472
2020-07-17
Volume2020Issue:1Pages:15
Contribution Rank1
Abstract

This paper addresses the traditional issue of restoring a high-resolution (HR) facial image from a low-resolution (LR) counterpart. Current state-of-the-art super-resolution (SR) methods commonly adopt the convolutional neural networks to learn a non-linear complex mapping between paired LR and HR images. They discriminate local patterns expressed by the neighboring pixels along the planar directions but ignore the intrinsic 3D proximity including the depth map. As a special case of general images, the face has limited geometric variations, which we believe that the relevant depth map can be learned and used to guide the face SR task. Motivated by it, we design a network including two branches: one for auxiliary depth map estimation and the other for the main SR task. Adaptive geometric features are further learned from the depth map and used to modulate the mid-level features of the SR branch. The whole network is implemented in an end-to-end trainable manner under the extra supervision of depth map. The supervisory depth map is either a paired one from RGB-D scans or a reconstructed one by a 3D prior model of faces. The experiments demonstrate the effectiveness of the proposed method and achieve improved performance over the state of the arts.

KeywordConvolutional neural networks Depth map Face super-resolution
DOI10.1186/s13638-020-01760-y
WOS KeywordCOMPUTATION OFFLOADING METHOD ; FACE RECOGNITION ; NETWORK
Indexed BySCI
Language英语
Funding ProjectNational Natural Science Foundation of China[61571438] ; National Key R&D Program of China[2017YFC0803505] ; National Key R&D Program of China[2017YFC0803505] ; National Natural Science Foundation of China[61571438]
WOS Research AreaEngineering ; Telecommunications
WOS SubjectEngineering, Electrical & Electronic ; Telecommunications
WOS IDBMC:10.1186/s13638-020-01760-y
PublisherSpringer International Publishing
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/40133
Collection智能制造技术与系统研究中心_多维数据分析
Corresponding AuthorHu, Xiyuan
Affiliation1.Institute of Automation, Chinese Academy of Sciences
2.University of Chinese Academy of Sciences
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Corresponding Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Fan, Zhenfeng,Hu, Xiyuan,Chen, Chen,et al. Facial image super-resolution guided by adaptive geometric features[J]. EURASIP Journal on Wireless Communications and Networking,2020,2020(1):15.
APA Fan, Zhenfeng,Hu, Xiyuan,Chen, Chen,Wang, Xiaolian,&Peng, Silong.(2020).Facial image super-resolution guided by adaptive geometric features.EURASIP Journal on Wireless Communications and Networking,2020(1),15.
MLA Fan, Zhenfeng,et al."Facial image super-resolution guided by adaptive geometric features".EURASIP Journal on Wireless Communications and Networking 2020.1(2020):15.
Files in This Item: Download All
File Name/Size DocType Version Access License
fan2020.pdf(4900KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Fan, Zhenfeng]'s Articles
[Hu, Xiyuan]'s Articles
[Chen, Chen]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fan, Zhenfeng]'s Articles
[Hu, Xiyuan]'s Articles
[Chen, Chen]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fan, Zhenfeng]'s Articles
[Hu, Xiyuan]'s Articles
[Chen, Chen]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: fan2020.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.