CASIA OpenIR  > 智能感知与计算
DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification
Zhang, Shu1; He, Ran1,2; Sun, Zhenan1,2; Tan, Tieniu1,2
AbstractMeshFace photos have been widely used in many Chinese business organizations to protect ID face photos from being misused. The occlusions incurred by random meshes severely degenerate the performance of face verification systems, which raises the MeshFace verification problem between MeshFace and daily photos. Previous methods cast this problem as a typical low-level vision problem, i.e., blind inpainting. They recover perceptually pleasing clear ID photos from MeshFaces by enforcing pixel level similarity between the recovered ID images and the ground-truth clear ID images and then perform face verification on them. Essentially, face verification is conducted on a compact feature space rather than the image pixel space. Therefore, this paper argues that pixel level similarity and feature level similarity jointly offer the key to improve the verification performance. Based on this insight, we offer a novel feature oriented blind face inpainting framework. Specifically, we implement this by establishing a novel DeMeshNet, which consists of three parts. The first part addresses blind inpainting of the MeshFaces by implicitly exploiting extra supervision from the occlusion position to enforce pixel level similarity. The second part explicitly enforces a feature level similarity in the compact feature space, which can explore informative supervision from the feature space to produce better inpainting results for verification. The last part copes with face alignment within the net via a customized spatial transformer module when extracting deep facial features. All three parts are implemented within an end-to-end network that facilitates efficient optimization. Extensive experiments on two MeshFace data sets demonstrate the effectiveness of the proposed DeMeshNet as well as the insight of this paper.
KeywordMeshface Face Verification Blind Inpainting Deep Learning Demeshnet Spatial Transformer
WOS HeadingsScience & Technology ; Technology
Indexed BySCI
Funding OrganizationNational Natural Science Foundation of China(61622310 ; Beijing Municipal Science and Technology Commission(Z161100000216144) ; State Key Development Program(2016YFB1001001) ; 61473289)
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS IDWOS:000418723000007
Citation statistics
Cited Times:31[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Affiliation1.Chinese Acad Sci, Ctr Res Intelligent Percept & Comp, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China
2.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing 100190, Peoples R China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Zhang, Shu,He, Ran,Sun, Zhenan,et al. DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification[J]. IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,2018,13(3):637-647.
APA Zhang, Shu,He, Ran,Sun, Zhenan,&Tan, Tieniu.(2018).DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification.IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,13(3),637-647.
MLA Zhang, Shu,et al."DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification".IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 13.3(2018):637-647.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhang, Shu]'s Articles
[He, Ran]'s Articles
[Sun, Zhenan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhang, Shu]'s Articles
[He, Ran]'s Articles
[Sun, Zhenan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhang, Shu]'s Articles
[He, Ran]'s Articles
[Sun, Zhenan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

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