Institute of Computing Technology, Chinese Academy IR
Deformable face net for pose invariant face recognition | |
He, Mingjie1,2; Zhang, Jie1; Shan, Shiguang1,2,3; Kan, Meina1; Chen, Xilin1,2 | |
2020-04-01 | |
发表期刊 | PATTERN RECOGNITION |
ISSN | 0031-3203 |
卷号 | 100页码:11 |
摘要 | Unconstrained face recognition still remains a challenging task due to various factors such as pose, expression, illumination, partial occlusion, etc. In particular, the most significant appearance variations are stemmed from poses which leads to severe performance degeneration. In this paper, we propose a novel Deformable Face Net (DFN) to handle the pose variations for face recognition. The deformable convolution module attempts to simultaneously learn face recognition oriented alignment and identity-preserving feature extraction. The displacement consistency loss (DCL) is proposed as a regularization term to enforce the learnt displacement fields for aligning faces to be locally consistent both in the orientation and amplitude since faces possess strong structure. Moreover, the identity consistency loss (ICL) and the pose-triplet loss (PTL) are designed to minimize the intra-class feature variation caused by different poses and maximize the inter-class feature distance under the same poses. The proposed DFN can effectively handle pose invariant face recognition (PIFR). Extensive experiments show that the proposed DFN outperforms the state-of-the-art methods, especially on the datasets with large poses. (C) 2019 Published by Elsevier Ltd. |
关键词 | Pose-invariant face recognition Displacement consistency loss Pose-triplet loss |
DOI | 10.1016/j.patcog.2019.107113 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Key R&D Program of China[2017YFA070080 0] ; Natural Science Foundation of China[61806188 and61772496] |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:000533530800016 |
出版者 | ELSEVIER SCI LTD |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://119.78.100.204/handle/2XEOYT63/15362 |
专题 | 中国科学院计算技术研究所期刊论文_英文 |
通讯作者 | Shan, Shiguang |
作者单位 | 1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Chinese Acad Sci CAS, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.Peng Cheng Lab, Shenzhen 518055, Peoples R China |
推荐引用方式 GB/T 7714 | He, Mingjie,Zhang, Jie,Shan, Shiguang,et al. Deformable face net for pose invariant face recognition[J]. PATTERN RECOGNITION,2020,100:11. |
APA | He, Mingjie,Zhang, Jie,Shan, Shiguang,Kan, Meina,&Chen, Xilin.(2020).Deformable face net for pose invariant face recognition.PATTERN RECOGNITION,100,11. |
MLA | He, Mingjie,et al."Deformable face net for pose invariant face recognition".PATTERN RECOGNITION 100(2020):11. |
条目包含的文件 | 条目无相关文件。 |
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