《计算机应用研究》|Application Research of Computers

基于深度学习的面部修复技术综述

Survey of facial completion techniques based on deep learning

免费全文下载 (已被下载 次)  
获取PDF全文
作者 刘颖,佘建初,公衍超,卢津,王富平,林庆帆,李莹华
机构 1.电子信息现场勘验应用技术公安部重点实验室,西安 710121;2.陕西省无线通信与信息处理技术国际合作研究中心,西安 710121;3.西安邮电大学 图像与信息处理研究所,西安 710121;4.新加坡Silicon Vision有限公司,新加坡 787820
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)01-002-0009-06
DOI 10.19734/j.issn.1001-3695.2019.10.0583
摘要 传统图像修复算法在修复区域涉及复杂非重复结构(如面部)时,不能准确捕捉到高级语义。近三年来基于深度学习的方法被应用于图像修复中,其修复结果的结构相似性较传统方法提高了10%以上。首先阐述了面部修复技术的研究发展历程,主要介绍了基于深度学习的面部修复算法,将其分为无监督和有监督两大类方法,在每一类中重点对近年来涌现的各种面部修复算法进行分析和总结;然后归纳了当前主流的六类图像数据集,以及算法性能评价指标;最后讨论了面部修复技术的未来研究方向。
关键词 面部图像修复; 深度学习; 生成对抗网络; 卷积神经网络
基金项目 国家自然科学基金资助项目(61801381)
陕西省国际合作交流项目(2018KW-003)
本文URL http://www.arocmag.com/article/01-2021-01-002.html
英文标题 Survey of facial completion techniques based on deep learning
作者英文名 Liu Ying, She Jianchu, Gong Yanchao, Lu Jin, Wang Fuping, Lim Kengpang, Li Yinghua
机构英文名 1.Key Laboratory of Electronic Information Application Technology for Crime Scene Investigation,Ministry of Public Security,Xi'an 710121,China;2.International Cooperation Research Center for Wireless Communication & Information Processing Technology,Xi'an 710121,China;3.Center for Image & Information Processing,Xi'an University of Posts & Telecommunications,Xi'an 710121,China;4.Xsecpro Pte.Ltd. ,Singapore 787820,Singapore
英文摘要 Traditional image inpainting algorithms cannot accurately capture high-level semantics when regions involve complex non-repetitive structures, such as faces. In the past three years, the method based on deep learning has been applied to image inpainting, and the structural similarity of the repair results had increased by more than 10% compared with the traditional methods. This paper firstly expounded the research and development process of facial completion technologies, mainly introduced the face repair algorithm based on deep learning, which was divided into unsupervised and supervised categories. In each category, it focused on the various facial completion algorithm ideas that had emerged in recent years. Then it summarized the six mainstream types of image datasets. Two evaluation indexes were summarized in this paper. Finally, it discussed the future research direction of facial completion technique.
英文关键词 facial image completion; deep learning; generative adversarial network(GAN); convolutional neural network(CNN)
参考文献 查看稿件参考文献
 
收稿日期 2019/10/17
修回日期 2019/12/11
页码 9-14
中图分类号 TP391.4
文献标志码 A