Title
Frame Size Reduction for Foreground Detection in Video Sequences.
Abstract
A frame resolution reduction framework to reduce the computational load and improve the foreground detection in video sequences is presented in this work. The proposed framework consists of three different stages. Firstly, the original video frame is downsampled using a specific interpolation function. Secondly, a foreground detection of the reduced video frame is performed by a probabilistic background model called MFBM. Finally, the class probabilities for the reduced video frame are upsampled using a bicubic interpolation to estimate the class probabilities of the original frame. Experimental results applied to standard benchmark video sequences demonstrate the goodness of our proposal.
Year
DOI
Venue
2016
10.1007/978-3-319-44636-3_1
ADVANCES IN ARTIFICIAL INTELLIGENCE, CAEPIA 2016
Keywords
DocType
Volume
Foreground detection,Video size reduction,Interpolation techniques
Conference
9868
ISSN
Citations 
PageRank 
0302-9743
0
0.34
References 
Authors
5
5