Online background subtraction with freely moving cameras using different motion boundaries

We propose a method for online background subtraction from a successive-frame video captured using a freely moving camera. Our method exploits a technique of interactive image segmentation with seeds (the subsets of pixels marked as “foreground” and “background”). The key novelty of our method is to...

Full description

Saved in:

Bibliographic Details
Published in:	Image and vision computing Vol. 76; pp. 76 - 92
Main Authors:	Sugimura, Daisuke, Teshima, Fumihiro, Hamamoto, Takayuki
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 01-08-2018
Subjects:	Freely moving camera Interactive image segmentation Motion boundary Online background subtraction Seeds estimation Online background subtraction Motion boundary Interactive image segmentation Seeds estimation Freely moving camera
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We propose a method for online background subtraction from a successive-frame video captured using a freely moving camera. Our method exploits a technique of interactive image segmentation with seeds (the subsets of pixels marked as “foreground” and “background”). The key novelty of our method is to automatically estimate the seeds by exploiting two different motion boundaries that are respectively computed using the magnitude and direction of the flow field. The magnitude of flow field is likely to be useful in differentiating the foreground and background motions when the moving objects and the camera make a movement towards the same direction. In contrast, the direction of flow field helps in discriminating the observed motions when the amount of displacement of the moving objects and the camera is the same. By adaptively exploiting the advantages of these different motion boundaries, our method enables to estimate the reliable foreground/background seeds. With the estimated seeds, our method performs accurate background subtraction even when the complex camera movements (e.g., large pan-tilt-zoom, rotation) are made. Our experiments demonstrate the effectiveness of our method using public dataset and other real image sequences. [Display omitted] •Online background subtraction with a freely-moving camera is proposed.•Our method utilizes interactive segmentation with seeds for background subtraction.•Automatic seeds estimation using two different motion boundaries is proposed.•Segmentation using the estimated seeds extracts foreground candidates.•Foregrounds are estimated from the candidates based on spatio-temporal consistency.
ISSN:	0262-8856 1872-8138
DOI:	10.1016/j.imavis.2018.06.003