Protecting Your Faces: MeshFaces Generation and Removal via High-order Relation-preserving CycleGAN | |
Zhihang Li1![]() ![]() ![]() | |
2018 | |
会议名称 | IAPR International Conference on Biometrics |
会议日期 | 2018 |
会议地点 | Queensland, Australia |
摘要 | Protecting person’s face photos from being mis- used has been an important issue as the rapid development of ubiquitous face sensors. MeshFaces provide a simple yet 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. |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/21093 |
专题 | 模式识别实验室 |
通讯作者 | Ran He(赫然) |
作者单位 | 1.National Laboratory of Pattern Recognition, CASIA 2.College of Information Engineering, Capital Normal University |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Zhihang Li,Yibo Hu,Man Zhang,et al. Protecting Your Faces: MeshFaces Generation and Removal via High-order Relation-preserving CycleGAN[C],2018. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
ICB.pdf(1955KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论