Adaptive background model registration for moving cameras

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We propose a framework for adaptively registering background models with an image for background subtraction with moving cameras. Existing methods search for a background model using a fixed window size, to suppress the number of false positives when detecting the foreground. However, these approaches result in many false negatives because they may use inappropriate window sizes. The appropriate size depends on various factors of the target scenes. To suppress false detections, we propose adaptively controlling the method parameters, which are typically determined heuristically. More specifically, the search window size for background registration and the foreground detection threshold are automatically determined using the re-projection error computed by the homography based camera motion estimate. Our method is based on the fact that the error at a pixel is low if it belongs to background and high if it does not. We quantitatively confirmed that the proposed framework improved the background subtraction accuracy when applied to images from moving cameras in various public datasets.

Original languageEnglish
Pages (from-to)86-95
Number of pages10
JournalPattern Recognition Letters
Volume96
DOIs
Publication statusPublished - Sep 1 2017

Fingerprint

Cameras
Pixels

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Adaptive background model registration for moving cameras. / Minematsu, Tsubasa; Uchiyama, Hideaki; Shimada, Atsushi; Nagahara, Hajime; Taniguchi, Rin-Ichiro.

In: Pattern Recognition Letters, Vol. 96, 01.09.2017, p. 86-95.

Research output: Contribution to journalArticle

@article{9e0b60bb603f4060bce787afa9a15004,
title = "Adaptive background model registration for moving cameras",
abstract = "We propose a framework for adaptively registering background models with an image for background subtraction with moving cameras. Existing methods search for a background model using a fixed window size, to suppress the number of false positives when detecting the foreground. However, these approaches result in many false negatives because they may use inappropriate window sizes. The appropriate size depends on various factors of the target scenes. To suppress false detections, we propose adaptively controlling the method parameters, which are typically determined heuristically. More specifically, the search window size for background registration and the foreground detection threshold are automatically determined using the re-projection error computed by the homography based camera motion estimate. Our method is based on the fact that the error at a pixel is low if it belongs to background and high if it does not. We quantitatively confirmed that the proposed framework improved the background subtraction accuracy when applied to images from moving cameras in various public datasets.",
author = "Tsubasa Minematsu and Hideaki Uchiyama and Atsushi Shimada and Hajime Nagahara and Rin-Ichiro Taniguchi",
year = "2017",
month = "9",
day = "1",
doi = "10.1016/j.patrec.2017.03.010",
language = "English",
volume = "96",
pages = "86--95",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",

}

TY - JOUR

T1 - Adaptive background model registration for moving cameras

AU - Minematsu, Tsubasa

AU - Uchiyama, Hideaki

AU - Shimada, Atsushi

AU - Nagahara, Hajime

AU - Taniguchi, Rin-Ichiro

PY - 2017/9/1

Y1 - 2017/9/1

N2 - We propose a framework for adaptively registering background models with an image for background subtraction with moving cameras. Existing methods search for a background model using a fixed window size, to suppress the number of false positives when detecting the foreground. However, these approaches result in many false negatives because they may use inappropriate window sizes. The appropriate size depends on various factors of the target scenes. To suppress false detections, we propose adaptively controlling the method parameters, which are typically determined heuristically. More specifically, the search window size for background registration and the foreground detection threshold are automatically determined using the re-projection error computed by the homography based camera motion estimate. Our method is based on the fact that the error at a pixel is low if it belongs to background and high if it does not. We quantitatively confirmed that the proposed framework improved the background subtraction accuracy when applied to images from moving cameras in various public datasets.

AB - We propose a framework for adaptively registering background models with an image for background subtraction with moving cameras. Existing methods search for a background model using a fixed window size, to suppress the number of false positives when detecting the foreground. However, these approaches result in many false negatives because they may use inappropriate window sizes. The appropriate size depends on various factors of the target scenes. To suppress false detections, we propose adaptively controlling the method parameters, which are typically determined heuristically. More specifically, the search window size for background registration and the foreground detection threshold are automatically determined using the re-projection error computed by the homography based camera motion estimate. Our method is based on the fact that the error at a pixel is low if it belongs to background and high if it does not. We quantitatively confirmed that the proposed framework improved the background subtraction accuracy when applied to images from moving cameras in various public datasets.

UR - http://www.scopus.com/inward/record.url?scp=85016473841&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016473841&partnerID=8YFLogxK

U2 - 10.1016/j.patrec.2017.03.010

DO - 10.1016/j.patrec.2017.03.010

M3 - Article

AN - SCOPUS:85016473841

VL - 96

SP - 86

EP - 95

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

ER -