保护你的脸:网格脸的生成和去除通过高阶关系保持CycleGAN

Zhihang Li, Yibo Hu, Man Zhang, Min Xu, R. He
{"title":"保护你的脸:网格脸的生成和去除通过高阶关系保持CycleGAN","authors":"Zhihang Li, Yibo Hu, Man Zhang, Min Xu, R. He","doi":"10.1109/ICB2018.2018.00020","DOIUrl":null,"url":null,"abstract":"Protecting person's face photos from being misused has been an important issue as the rapid development of ubiquitous face sensors. MeshFaces provide a simple and inexpensive way to protect facial photos and have been widely used in China. This paper treats MeshFace generation and removal as a dual learning problem and proposes a high-order relation-preserving CycleGAN framework to solve this problem. First, dual transformations between the distributions of MeshFaces and clean faces in pixel space are learned under the CycleGAN framework, which can efficiently utilize unpaired data. Then, a novel High-order Relation-preserving (HR) loss is imposed on CycleGAN to recover the finer texture details and generate much sharper images. Different from the L1 and L2 losses that result in image smoothness and blurry, the HR loss can better capture the appearance variation of MeshFaces and hence facilitates removal. Moreover, Identity Preserving loss is proposed to preserve both global and local identity information. Experimental results on three databases demonstrate that our approach is highly effective for MeshFace generation and removal.","PeriodicalId":130957,"journal":{"name":"2018 International Conference on Biometrics (ICB)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":"{\"title\":\"Protecting Your Faces: MeshFaces Generation and Removal via High-Order Relation-Preserving CycleGAN\",\"authors\":\"Zhihang Li, Yibo Hu, Man Zhang, Min Xu, R. He\",\"doi\":\"10.1109/ICB2018.2018.00020\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Protecting person's face photos from being misused has been an important issue as the rapid development of ubiquitous face sensors. MeshFaces provide a simple and inexpensive way to protect facial photos and have been widely used in China. This paper treats MeshFace generation and removal as a dual learning problem and proposes a high-order relation-preserving CycleGAN framework to solve this problem. First, dual transformations between the distributions of MeshFaces and clean faces in pixel space are learned under the CycleGAN framework, which can efficiently utilize unpaired data. Then, a novel High-order Relation-preserving (HR) loss is imposed on CycleGAN to recover the finer texture details and generate much sharper images. Different from the L1 and L2 losses that result in image smoothness and blurry, the HR loss can better capture the appearance variation of MeshFaces and hence facilitates removal. Moreover, Identity Preserving loss is proposed to preserve both global and local identity information. Experimental results on three databases demonstrate that our approach is highly effective for MeshFace generation and removal.\",\"PeriodicalId\":130957,\"journal\":{\"name\":\"2018 International Conference on Biometrics (ICB)\",\"volume\":\"8 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"11\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 International Conference on Biometrics (ICB)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICB2018.2018.00020\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 International Conference on Biometrics (ICB)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICB2018.2018.00020","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 11

摘要

随着人脸传感器的迅速发展,保护人脸照片不被滥用已经成为一个重要的问题。MeshFaces提供了一种简单而廉价的面部照片保护方法,在中国得到了广泛的应用。本文将MeshFace的生成和去除视为一个双重学习问题,并提出了一个高阶保持关系的CycleGAN框架来解决这个问题。首先,在CycleGAN框架下学习了MeshFaces和clean faces在像素空间的分布对偶变换,有效地利用了未配对数据;然后,在CycleGAN上施加一种新的高阶关系保持(High-order relationship -preserving, HR)损失,以恢复更精细的纹理细节,生成更清晰的图像。与L1和L2损失导致图像平滑和模糊不同,HR损失可以更好地捕捉MeshFaces的外观变化,从而便于去除。此外,还提出了身份保留损失,以同时保存全局和局部身份信息。在三个数据库上的实验结果表明,我们的方法对MeshFace的生成和去除是非常有效的。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Protecting Your Faces: MeshFaces Generation and Removal via High-Order Relation-Preserving CycleGAN
Protecting person's face photos from being misused has been an important issue as the rapid development of ubiquitous face sensors. MeshFaces provide a simple and inexpensive way to protect facial photos and have been widely used in China. This paper treats MeshFace generation and removal as a dual learning problem and proposes a high-order relation-preserving CycleGAN framework to solve this problem. First, dual transformations between the distributions of MeshFaces and clean faces in pixel space are learned under the CycleGAN framework, which can efficiently utilize unpaired data. Then, a novel High-order Relation-preserving (HR) loss is imposed on CycleGAN to recover the finer texture details and generate much sharper images. Different from the L1 and L2 losses that result in image smoothness and blurry, the HR loss can better capture the appearance variation of MeshFaces and hence facilitates removal. Moreover, Identity Preserving loss is proposed to preserve both global and local identity information. Experimental results on three databases demonstrate that our approach is highly effective for MeshFace generation and removal.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信