DocumentCode :
179065
Title :
Practical ReProCS for separating sparse and low-dimensional signal sequences from their sum — Part 1
Author :
Han Guo ; Chenlu Qiu ; Vaswani, Namrata
Author_Institution :
ECE Dept., Iowa State Univ., Ames, IA, USA
fYear :
2014
fDate :
4-9 May 2014
Firstpage :
4161
Lastpage :
4165
Abstract :
This paper designs and evaluates a practical algorithm, called Prac-ReProCS, for recovering a time sequence of sparse vectors St and a time sequence of dense vectors Lt from their sum, Mt := St + Lt, when any subsequence of the Lt´s lies in a slowly changing low-dimensional subspace. A key application where this problem occurs is in video layering where the goal is to separate a video sequence into a slowly changing background sequence and a sparse foreground sequence that consists of one or more moving regions/objects. Prac-ReProCS is the practical analog of its theoretical counterpart that was studied in our recent work.
Keywords :
image sequences; source separation; vectors; Prac-ReProCS; background sequence; dense vector time sequence recovery; low-dimensional signal sequence; sparse foreground sequence; sparse separation; sparse vector time sequence recovery; video layering; video sequence separation; Compressed sensing; Matrix decomposition; Noise; Principal component analysis; Robustness; Sparse matrices; Vectors; compressed sensing; robust PCA; robust matrix completion; sparse recovery;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
Conference_Location :
Florence
Type :
conf
DOI :
10.1109/ICASSP.2014.6854385
Filename :
6854385
Link To Document :
بازگشت