Denoising-Based Turbo Message Passing for Compressed Video Background Subtraction

doi:10.1109/TIP.2021.3055063

KMS > 科道书院

	Denoising-Based Turbo Message Passing for Compressed Video Background Subtraction
	Zhipeng Xue 1; Xiaojun Yuan 1; Yang Yang2
	2021
发表期刊	IEEE TRANSACTIONS ON IMAGE PROCESSING
ISSN	1057-7149
EISSN	1941-0042
卷号	30 页码:2682-2696
发表状态	已发表
DOI	10.1109/TIP.2021.3055063
摘要	In this paper, we consider the compressed video background subtraction problem that separates the background and foreground of a video from its compressed measurements. The background of a video usually lies in a low dimensional space and the foreground is usually sparse. More importantly, each video frame is a natural image that has textural patterns. By exploiting these properties, we develop a message passing algorithm termed offline denoising-based turbo message passing (DTMP). We show that these structural properties can be efficiently handled by the existing denoising techniques under the turbo message passing framework. We further extend the DTMP algorithm to the online scenario where the video data is collected in an online manner. The extension is based on the similarity/continuity between adjacent video frames. We adopt the optical flow method to refine the estimation of the foreground. We also adopt the sliding window based background estimation to reduce complexity. By exploiting the Gaussianity of messages, we develop the state evolution to characterize the per-iteration performance of offline and online DTMP. Comparing to the existing algorithms, DTMP can work at much lower compression rates, and can subtract the background successfully with a lower mean squared error and better visual quality for both offline and online compressed video background subtraction.
关键词	Message passing Image coding Approximation algorithms Sparse matrices Estimation Optical imaging Neural networks Background subtraction compressive measurement message passing turbo principle
URL	查看原文
收录类别	SCI ; EI ; SCIE
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000616314200017
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
原始文献类型	Article
来源库	IEEE
引用统计
文献类型	期刊论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/125919
专题	科道书院信息科学与技术学院_PI研究组_杨旸组
作者单位	1.Center for Intelligent Networking and Communications (CINC), University of Electronic Science and Technology of China, Chengdu, China 2.Shanghai Institute of Fog Computing Technology (SHIFT), ShanghaiTech University, Shanghai, China
推荐引用方式 GB/T 7714	Zhipeng Xue,Xiaojun Yuan,Yang Yang. Denoising-Based Turbo Message Passing for Compressed Video Background Subtraction[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2021,30:2682-2696.
APA	Zhipeng Xue,Xiaojun Yuan,&Yang Yang.(2021).Denoising-Based Turbo Message Passing for Compressed Video Background Subtraction.IEEE TRANSACTIONS ON IMAGE PROCESSING,30,2682-2696.
MLA	Zhipeng Xue,et al."Denoising-Based Turbo Message Passing for Compressed Video Background Subtraction".IEEE TRANSACTIONS ON IMAGE PROCESSING 30(2021):2682-2696.