Improving video foreground segmentation with an object-like pool

doi:10.1117/1.JEI.24.2.023034

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

	Improving video foreground segmentation with an object-like pool
	Cheng, Xiaoliu 1,2; Lv, Wei 2; Liu, Huawei 1,2; You, Xing 2; Li, Baoqing 2; Yuan, Xiaobing 2
刊名	JOURNAL OF ELECTRONIC IMAGING
	2015-03-01
卷号	24 期号:2 页码:8
关键词	video foreground segmentation object-like pool unsupervised unlabeled conditional random field probabilistic superpixels
ISSN号	1017-9909
DOI	10.1117/1.JEI.24.2.023034
英文摘要	Foreground segmentation in video frames is quite valuable for object and activity recognition, while the existing approaches often demand training data or initial annotation, which is expensive and inconvenient. We propose an automatic and unsupervised method of foreground segmentation given an unlabeled and short video. The pixel-level optical flow and binary mask features are converted into the normal probabilistic superpixels, therefore, they are adaptable to build the superpixel-level conditional random field which aims to label the foreground and background. We exploit the fact that the appearance and motion features of the moving object are temporally and spatially coherent in general, to construct an object-like pool and background-like pool via the previous segmented results. The continuously updated pools can be regarded as the "prior" knowledge of the current frame to provide a reliable way to learn the features of the object. Experimental results demonstrate that our approach exceeds the current methods, both qualitatively and quantitatively. (C) The Authors.
资助项目	National Natural Science Foundation of China[14ZR1447200]
WOS研究方向	Engineering ; Optics ; Imaging Science & Photographic Technology
语种	英语
出版者	IS&T & SPIE
WOS记录号	WOS:000354873600034
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/9685]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	You, Xing
作者单位	1.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Shanghai Inst Microsyst & Informat Technol, Wireless Sensor Network Lab, Shanghai 200050, Peoples R China
推荐引用方式 GB/T 7714	Cheng, Xiaoliu,Lv, Wei,Liu, Huawei,et al. Improving video foreground segmentation with an object-like pool[J]. JOURNAL OF ELECTRONIC IMAGING,2015,24(2):8.
APA	Cheng, Xiaoliu,Lv, Wei,Liu, Huawei,You, Xing,Li, Baoqing,&Yuan, Xiaobing.(2015).Improving video foreground segmentation with an object-like pool.JOURNAL OF ELECTRONIC IMAGING,24(2),8.
MLA	Cheng, Xiaoliu,et al."Improving video foreground segmentation with an object-like pool".JOURNAL OF ELECTRONIC IMAGING 24.2(2015):8.