Knowledge Commons of Institute of Automation,CAS
Learning Meta Face Recognition in Unseen Domains | |
Guo JZ(郭建珠)1,2; Zhu XY(朱翔昱)1,2; Zhao CX(赵晨旭)3; Cao D(曹冬)1,2; Lei Z(雷震)1,2; Li ZQ(李子青)4 | |
2020-06 | |
会议名称 | IEEE/CVF Conference on Computer Vision and Pattern Recognition |
页码 | 6162-6171 |
会议日期 | June 13-19, 2020 |
会议地点 | Seattle, WA, USA |
出版者 | IEEE |
摘要 | Face recognition systems are usually faced with unseen domains in real-world applications and show unsatisfactory performance due to their poor generalization. For example, a well-trained model on webface data cannot deal with the ID vs. Spot task in surveillance scenario. In this paper, we aim to learn a generalized model that can directly handle new unseen domains without any model updating. To this end, we propose a novel face recognition method via meta-learning named Meta Face Recognition (MFR). MFR synthesizes the source/target domain shift with a meta-optimization objective, which requires the model to learn effective representations not only on synthesized source domains but also on synthesized target domains. Specifically, we build domain-shift batches through a domain-level sampling strategy and get back-propagated gradients/meta-gradients on synthesized source/target domains by optimizing multi-domain distributions. The gradients and meta-gradients are further combined to update the model to improve generalization. Besides, we propose two benchmarks for generalized face recognition evaluation. Experiments on our benchmarks validate the generalization of our method compared to several baselines and other state-of-the-arts. The proposed benchmarks and code will be available at https://github.com/cleardusk/MFR. |
DOI | 10.1109/CVPR42600.2020.00620 |
收录类别 | EI |
语种 | 英语 |
七大方向——子方向分类 | 生物特征识别 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/44370 |
专题 | 多模态人工智能系统全国重点实验室_生物识别与安全技术 智能感知与计算研究中心 |
通讯作者 | Lei Z(雷震) |
作者单位 | 1.中国科学院自动化所 2.中国科学院大学 3.明略科技 4.西湖大学 |
推荐引用方式 GB/T 7714 | Guo JZ,Zhu XY,Zhao CX,et al. Learning Meta Face Recognition in Unseen Domains[C]:IEEE,2020:6162-6171. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Guo_Learning_Meta_Fa(3373KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
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