A deep attention-based ensemble network for real-time face hallucination

doi:10.1007/s11554-020-01009-3

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	A deep attention-based ensemble network for real-time face hallucination
	Liu, Dongdong 1; Chen, Jincai 1,2; Huang, Zhenxing 1; Zeng, Ni 3; Lu, Ping 2; Yang, Lin 4; Wang, Haofeng 4; Kou, Jinqiao 4; Wu, Min 4
刊名	JOURNAL OF REAL-TIME IMAGE PROCESSING
	2020-08-17
页码	11
关键词	Face hallucination Attention mechanism Residual learning Ensemble model
ISSN号	1861-8200
DOI	10.1007/s11554-020-01009-3
英文摘要	Face hallucination (FH) aims to reconstruct high-resolution faces from low-resolution face inputs, making it significant to other face-related tasks. Different from general super resolution issue, it often requires facial priors other than general extracted features thus leading to fusion of more than one kind of feature. The existing CNN-based FH methods often fuse different features indiscriminately which may introduce noises. Also the latent relations among different features which may be useful are taken into less consideration. To address the above issues, we propose an end-to-end deep ensemble network which aggregates three extraction sub-nets in attention-based manner. In our ensemble strategy, both relations among different features and inter-dependencies among different channels are dug out through the exploitation of spatial attention and channel attention. And for the diversity of extracted features, we aggregate three different sub-nets, which are the basic sub-net for basic features, the auto-encoder sub-net for facial shape priors and the dense residual attention sub-net for fine-grained texture features. Conducted ablation studies and experimental results show that our method achieves effectiveness not only in PSNR (Peak Signal to Noise Ratio) and SSIM (Structural Similarity Index) metrics but more importantly in clearer details within both key facial areas and whole range. Also results show that our method achieves real-time hallucinating faces by generating one image in 0.0237s.
资助项目	National Natural Science Foundation of China[61672246] ; National Natural Science Foundation of China[61272068] ; Fundamental Research Funds for the Central Universities[HUST:2016YXMS018]
WOS研究方向	Computer Science ; Engineering ; Imaging Science & Photographic Technology
语种	英语
出版者	SPRINGER HEIDELBERG
WOS记录号	WOS:000560272800001
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/15823]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Chen, Jincai
作者单位	1.Huazhong Univ Sci & Technol, Wuhan Natl Lab Optoelect, Wuhan, Peoples R China 2.Huazhong Univ Sci & Technol, Sch Compter Sci & Technol, Wuhan, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing Key Lab Mobile Comp & Pervas Devices, Beijing, Peoples R China 4.Beijing Inst Comp Technol & Applicat, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Liu, Dongdong,Chen, Jincai,Huang, Zhenxing,et al. A deep attention-based ensemble network for real-time face hallucination[J]. JOURNAL OF REAL-TIME IMAGE PROCESSING,2020:11.
APA	Liu, Dongdong.,Chen, Jincai.,Huang, Zhenxing.,Zeng, Ni.,Lu, Ping.,...&Wu, Min.(2020).A deep attention-based ensemble network for real-time face hallucination.JOURNAL OF REAL-TIME IMAGE PROCESSING,11.
MLA	Liu, Dongdong,et al."A deep attention-based ensemble network for real-time face hallucination".JOURNAL OF REAL-TIME IMAGE PROCESSING (2020):11.