Knowledge Commons of Institute of Automation,CAS
OS-LFFD: a light and fast face detector with Ommateum structure | |
Xu, Dezhong1; Wu, Lifang1; He, Yonghao1,2; Zhao, Qing1; Jian, Meng1; Yan, Junchi3; Zhao, Liang1 | |
发表期刊 | MULTIMEDIA TOOLS AND APPLICATIONS |
ISSN | 1380-7501 |
2020-07-13 | |
页码 | 20 |
通讯作者 | Zhao, Qing(zhaoqing1025@emails.bjut.edu.cn) |
摘要 | Face detection has been deployed on edge devices as the basis for face applications, but the devices cannot store large-scale models and have low computing power. The existing anchor-based face detection schemes cannot cover face images over a continuous size range, and their performance is not satisfactory. Obviously, good performances are accompanied by increased storage and lower speed. We find that the feature points in different layers correspond to a specific size range of RFs (receptive fields). According to the survey, the predictable range of RFs with the same size is the face on a continuous scale. Therefore, we argue that RFs are inherent anchors. A Light and Fast Face Detector with an Ommateum Structure (OS-LFFD) is proposed in this paper. By analyzing the correlation between the effective receptive field (ERF) and face sizes, a 4-branch network is designed to cover the objective range of face sizes. Each branch involves an ommateum block with a similar structure and shared parameters. It reduces the number of model parameters (8 M), which makes it much smaller than most face detectors. Experiments on the popular benchmarks WIDER FACE and FDDB using multiple hardware platforms demonstrate that the proposed scheme can considerably balance the accuracy and running speed. |
关键词 | Edge devices Face detector Effective receptive field Ommateum block |
DOI | 10.1007/s11042-020-09143-7 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Natural Science Foundation of China[61976010] ; National Natural Science Foundation of China[61702022] ; National Natural Science Foundation of China[61802011] ; Beijing Municipal Education Committee Science Foundation[KM201910005024] ; China Postdoctoral Science Foundation[2018M640033] ; Beijing Excellent Young Talent Cultivation Project[2017000020124G075] ; Beijing University of Technology Ri xin Cultivation Project |
项目资助者 | National Natural Science Foundation of China ; Beijing Municipal Education Committee Science Foundation ; China Postdoctoral Science Foundation ; Beijing Excellent Young Talent Cultivation Project ; Beijing University of Technology Ri xin Cultivation Project |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Information Systems ; Computer Science, Software Engineering ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:000548128400005 |
出版者 | SPRINGER |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/40160 |
专题 | 多模态人工智能系统全国重点实验室_先进时空数据分析与学习 |
通讯作者 | Zhao, Qing |
作者单位 | 1.Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Shanghai Jiao Tong Univ, Artificial Intelligence Inst, Dept Comp Sci & Engn, Shanghai 200240, Peoples R China |
推荐引用方式 GB/T 7714 | Xu, Dezhong,Wu, Lifang,He, Yonghao,et al. OS-LFFD: a light and fast face detector with Ommateum structure[J]. MULTIMEDIA TOOLS AND APPLICATIONS,2020:20. |
APA | Xu, Dezhong.,Wu, Lifang.,He, Yonghao.,Zhao, Qing.,Jian, Meng.,...&Zhao, Liang.(2020).OS-LFFD: a light and fast face detector with Ommateum structure.MULTIMEDIA TOOLS AND APPLICATIONS,20. |
MLA | Xu, Dezhong,et al."OS-LFFD: a light and fast face detector with Ommateum structure".MULTIMEDIA TOOLS AND APPLICATIONS (2020):20. |
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