Moving-object segmentation using a foreground history map

Sooyeong Kwak, Guntae Bae, Hyeran Byun

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

This paper describes a real-time foreground segmentation method in monocular video sequences for video teleconferencing. Background subtraction is widely used in foreground segmentation for static cameras. However, the results are usually not accurate enough for background substitution tasks. In this paper, we propose a novel strategy for fast and accurate foreground segmentation. The strategy consists of two steps: initial foreground segmentation and fine foreground segmentation. The key to our algorithm consists of two steps. In the first step, a moving object is roughly segmented using the background subtraction method. In order to update the initial foreground segmentation results in the second step, a region-based segmentation method and a foreground history map (FHM)-based segmentation representing the combination of temporal and spatial information were developed. The segmentation accuracy of the proposed algorithm was evaluated with respect to the ground truth, which was the manually cropped foreground. The experimental results showed that the proposed algorithm improved the accuracy of segmentation with respect to Horprasert's well-known algorithm.

Original languageEnglish
Pages (from-to)180-187
Number of pages8
JournalJournal of the Optical Society of America A: Optics and Image Science, and Vision
Volume27
Issue number2
DOIs
Publication statusPublished - 2010 Feb 1

Fingerprint

histories
subtraction
video conferencing
Teleconferencing
ground truth
Substitution reactions
Cameras
cameras
substitutes

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Computer Vision and Pattern Recognition

Cite this

@article{d5fc35ea04f44dec935b87afd32a51cc,
title = "Moving-object segmentation using a foreground history map",
abstract = "This paper describes a real-time foreground segmentation method in monocular video sequences for video teleconferencing. Background subtraction is widely used in foreground segmentation for static cameras. However, the results are usually not accurate enough for background substitution tasks. In this paper, we propose a novel strategy for fast and accurate foreground segmentation. The strategy consists of two steps: initial foreground segmentation and fine foreground segmentation. The key to our algorithm consists of two steps. In the first step, a moving object is roughly segmented using the background subtraction method. In order to update the initial foreground segmentation results in the second step, a region-based segmentation method and a foreground history map (FHM)-based segmentation representing the combination of temporal and spatial information were developed. The segmentation accuracy of the proposed algorithm was evaluated with respect to the ground truth, which was the manually cropped foreground. The experimental results showed that the proposed algorithm improved the accuracy of segmentation with respect to Horprasert's well-known algorithm.",
author = "Sooyeong Kwak and Guntae Bae and Hyeran Byun",
year = "2010",
month = "2",
day = "1",
doi = "10.1364/JOSAA.27.000180",
language = "English",
volume = "27",
pages = "180--187",
journal = "Journal of the Optical Society of America A: Optics and Image Science, and Vision",
issn = "1084-7529",
publisher = "The Optical Society",
number = "2",

}

Moving-object segmentation using a foreground history map. / Kwak, Sooyeong; Bae, Guntae; Byun, Hyeran.

In: Journal of the Optical Society of America A: Optics and Image Science, and Vision, Vol. 27, No. 2, 01.02.2010, p. 180-187.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Moving-object segmentation using a foreground history map

AU - Kwak, Sooyeong

AU - Bae, Guntae

AU - Byun, Hyeran

PY - 2010/2/1

Y1 - 2010/2/1

N2 - This paper describes a real-time foreground segmentation method in monocular video sequences for video teleconferencing. Background subtraction is widely used in foreground segmentation for static cameras. However, the results are usually not accurate enough for background substitution tasks. In this paper, we propose a novel strategy for fast and accurate foreground segmentation. The strategy consists of two steps: initial foreground segmentation and fine foreground segmentation. The key to our algorithm consists of two steps. In the first step, a moving object is roughly segmented using the background subtraction method. In order to update the initial foreground segmentation results in the second step, a region-based segmentation method and a foreground history map (FHM)-based segmentation representing the combination of temporal and spatial information were developed. The segmentation accuracy of the proposed algorithm was evaluated with respect to the ground truth, which was the manually cropped foreground. The experimental results showed that the proposed algorithm improved the accuracy of segmentation with respect to Horprasert's well-known algorithm.

AB - This paper describes a real-time foreground segmentation method in monocular video sequences for video teleconferencing. Background subtraction is widely used in foreground segmentation for static cameras. However, the results are usually not accurate enough for background substitution tasks. In this paper, we propose a novel strategy for fast and accurate foreground segmentation. The strategy consists of two steps: initial foreground segmentation and fine foreground segmentation. The key to our algorithm consists of two steps. In the first step, a moving object is roughly segmented using the background subtraction method. In order to update the initial foreground segmentation results in the second step, a region-based segmentation method and a foreground history map (FHM)-based segmentation representing the combination of temporal and spatial information were developed. The segmentation accuracy of the proposed algorithm was evaluated with respect to the ground truth, which was the manually cropped foreground. The experimental results showed that the proposed algorithm improved the accuracy of segmentation with respect to Horprasert's well-known algorithm.

UR - http://www.scopus.com/inward/record.url?scp=76749139583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76749139583&partnerID=8YFLogxK

U2 - 10.1364/JOSAA.27.000180

DO - 10.1364/JOSAA.27.000180

M3 - Article

VL - 27

SP - 180

EP - 187

JO - Journal of the Optical Society of America A: Optics and Image Science, and Vision

JF - Journal of the Optical Society of America A: Optics and Image Science, and Vision

SN - 1084-7529

IS - 2

ER -