Hierarchical Attention for Part-Aware Face Detection

doi:10.1007/s11263-019-01157-5

	Hierarchical Attention for Part-Aware Face Detection
	Wu, Shuzhe 1,2; Kan, Meina 1; Shan, Shiguang 1,2,3; Chen, Xilin 1,3
	2019-06-01
发表期刊	INTERNATIONAL JOURNAL OF COMPUTER VISION
ISSN	0920-5691
卷号	127 期号:6-7 页码:560-578
摘要	Expressive representations for characterizing face appearances are essential for accurate face detection. Due to different poses, scales, illumination, occlusion, etc, face appearances generally exhibit substantial variations, and the contents of each local region (facial part) vary from one face to another. Current detectors, however, particularly those based on convolutional neural networks, apply identical operations (e.g. convolution or pooling) to all local regions on each face for feature aggregation (in a generic sliding-window configuration), and take all local features as equally effective for the detection task. In such methods, not only is each local feature suboptimal due to ignoring region-wise distinctions, but also the overall face representations are semantically inconsistent. To address the issue, we design a hierarchical attention mechanism to allow adaptive exploration of local features. Given a face proposal, part-specific attention modeled as learnable Gaussian kernels is proposed to search for proper positions and scales of local regions to extract consistent and informative features of facial parts. Then face-specific attention predicted with LSTM is introduced to model relations between the local parts and adjust their contributions to the detection tasks. Such hierarchical attention leads to a part-aware face detector, which forms more expressive and semantically consistent face representations. Extensive experiments are performed on three challenging face detection datasets to demonstrate the effectiveness of our hierarchical attention and make comparisons with state-of-the-art methods.
关键词	Hierarchical attention Face detection Object detection Deformation Part-aware
DOI	10.1007/s11263-019-01157-5
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2017YFA0700800] ; Natural Science Foundation of China[61390511] ; Natural Science Foundation of China[61650202] ; Natural Science Foundation of China[61772496] ; Natural Science Foundation of China[61402443]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000468525900003
出版者	SPRINGER
引用统计	被引频次：52[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/4230
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, ICT, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.UCAS, Beijing 100049, Peoples R China 3.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Shanghai 200031, Peoples R China
推荐引用方式 GB/T 7714	Wu, Shuzhe,Kan, Meina,Shan, Shiguang,et al. Hierarchical Attention for Part-Aware Face Detection[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2019,127(6-7):560-578.
APA	Wu, Shuzhe,Kan, Meina,Shan, Shiguang,&Chen, Xilin.(2019).Hierarchical Attention for Part-Aware Face Detection.INTERNATIONAL JOURNAL OF COMPUTER VISION,127(6-7),560-578.
MLA	Wu, Shuzhe,et al."Hierarchical Attention for Part-Aware Face Detection".INTERNATIONAL JOURNAL OF COMPUTER VISION 127.6-7(2019):560-578.