Detecting Face with Densely Connected Face Proposal Network

doi:10.1016/j.neucom.2018.01.012

CASIA OpenIR > 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

	Detecting Face with Densely Connected Face Proposal Network
	Zhang, Shifeng1,2,3 ; Zhu, Xiangyu1,2,3 ; Lei, Zhen1,2,3 ; Wang, Xiaobo1,2,3 ; Shi, Hailin1,2,3 ; Li, Stan Z.1,2,3
发表期刊	NEUROCOMPUTING
	2018-04-05
卷号	284 页码:119-127
文章类型	Article
摘要	Accuracy and efficiency are two conflicting challenges for face detection, since effective models tend to be computationally prohibitive. To address these two conflicting challenges, our core idea is to shrink the input image and focus on detecting small faces. Reducing the image resolution can significantly improve the detection speed, but it also results in smaller faces that need to pay more attention. Specifically, we propose a novel face detector, dubbed the name Densely Connected Face Proposal Network (DCFPN), with high accuracy as well as CPU real-time speed. Firstly, we subtly design a lightweight-but-powerful fully convolution network with the consideration of efficiency and accuracy. Secondly, we present a dense anchor strategy and a scale-aware anchor matching scheme to improve the recall rate of small faces. Finally, a fair L1 loss is introduced to locate small faces well. As a consequence, our proposed method can detect faces at 30 FPS on a single 2.60 GHz CPU core and 250 FPS using a GPU for the VGA-resolution images. We achieve state-of-the-art performance on the common face detection benchmark datasets. (C) 2018 Elsevier B.V. All rights reserved.
关键词	Face Detection Small Face Region Proposal Network
WOS标题词	Science & Technology ; Technology
DOI	10.1016/j.neucom.2018.01.012
关键词[WOS]	WILD
收录类别	SCI
语种	英语
项目资助者	National Key Research and Development Plan(2016YFC0801002) ; Chinese National Natural Science Foundation(61473291 ; AuthenMetric RD Funds ; 61572501 ; 61502491 ; 61572536)
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000425883300013
引用统计	被引频次：14[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/21950
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
作者单位	1.Chinese Acad Sci, CBSR, Inst Automat, Beijing, Peoples R China 2.Chinese Acad Sci, NLPR, Inst Automat, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Zhang, Shifeng,Zhu, Xiangyu,Lei, Zhen,et al. Detecting Face with Densely Connected Face Proposal Network[J]. NEUROCOMPUTING,2018,284:119-127.
APA	Zhang, Shifeng,Zhu, Xiangyu,Lei, Zhen,Wang, Xiaobo,Shi, Hailin,&Li, Stan Z..(2018).Detecting Face with Densely Connected Face Proposal Network.NEUROCOMPUTING,284,119-127.
MLA	Zhang, Shifeng,et al."Detecting Face with Densely Connected Face Proposal Network".NEUROCOMPUTING 284(2018):119-127.

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
Neurocomputing_DCFPN（4200KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载
Neurocomputing_DCFPN（4200KB）			开放获取	--	浏览下载