The use of vanishing point for the classification of reflections from foreground mask in videos

László Havasi, Zoltán Szlávik, T. Szirányi

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Extraction of foreground is a basic task in surveillance video analysis. In most real cases, its performance is heavily based on the efficiency of shadow detection and on the analysis of lighting conditions and reflections caused by mirrors or other reflective surfaces. This correspondence is focused on the improvement of foreground extraction in the case of planar reflective surfaces. We show that the geometric model of a scene with a planar reflective surface is reduced to the estimation of vanishing-point for the case of an auto-epipolar (skew-symmetric) fundamental matrix. The correspondences for the vanishing-point estimation are extracted from motion statistics. The knowledge of the position of the vanishing point allows us to integrate the geometric model and the motion statistics into image foreground-extraction to separate foreground from reflections, and thus to achieve better performance. The experiments confirm the accuracy of the vanishing point and the improvement of the foreground image mask by removing reflected object parts.

Original languageEnglish
Pages (from-to)1366-1372
Number of pages7
JournalIEEE Transactions on Image Processing
Volume18
Issue number6
DOIs
Publication statusPublished - 2009

Fingerprint

Masks
Statistics
Lighting
Experiments

Keywords

  • Image and video processing
  • Robust estimation
  • Scene analysis

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

The use of vanishing point for the classification of reflections from foreground mask in videos. / Havasi, László; Szlávik, Zoltán; Szirányi, T.

In: IEEE Transactions on Image Processing, Vol. 18, No. 6, 2009, p. 1366-1372.

Research output: Contribution to journalArticle

@article{ac2e5326706f44c28b10b9d7686d97ec,
title = "The use of vanishing point for the classification of reflections from foreground mask in videos",
abstract = "Extraction of foreground is a basic task in surveillance video analysis. In most real cases, its performance is heavily based on the efficiency of shadow detection and on the analysis of lighting conditions and reflections caused by mirrors or other reflective surfaces. This correspondence is focused on the improvement of foreground extraction in the case of planar reflective surfaces. We show that the geometric model of a scene with a planar reflective surface is reduced to the estimation of vanishing-point for the case of an auto-epipolar (skew-symmetric) fundamental matrix. The correspondences for the vanishing-point estimation are extracted from motion statistics. The knowledge of the position of the vanishing point allows us to integrate the geometric model and the motion statistics into image foreground-extraction to separate foreground from reflections, and thus to achieve better performance. The experiments confirm the accuracy of the vanishing point and the improvement of the foreground image mask by removing reflected object parts.",
keywords = "Image and video processing, Robust estimation, Scene analysis",
author = "L{\'a}szl{\'o} Havasi and Zolt{\'a}n Szl{\'a}vik and T. Szir{\'a}nyi",
year = "2009",
doi = "10.1109/TIP.2009.2017137",
language = "English",
volume = "18",
pages = "1366--1372",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

TY - JOUR

T1 - The use of vanishing point for the classification of reflections from foreground mask in videos

AU - Havasi, László

AU - Szlávik, Zoltán

AU - Szirányi, T.

PY - 2009

Y1 - 2009

N2 - Extraction of foreground is a basic task in surveillance video analysis. In most real cases, its performance is heavily based on the efficiency of shadow detection and on the analysis of lighting conditions and reflections caused by mirrors or other reflective surfaces. This correspondence is focused on the improvement of foreground extraction in the case of planar reflective surfaces. We show that the geometric model of a scene with a planar reflective surface is reduced to the estimation of vanishing-point for the case of an auto-epipolar (skew-symmetric) fundamental matrix. The correspondences for the vanishing-point estimation are extracted from motion statistics. The knowledge of the position of the vanishing point allows us to integrate the geometric model and the motion statistics into image foreground-extraction to separate foreground from reflections, and thus to achieve better performance. The experiments confirm the accuracy of the vanishing point and the improvement of the foreground image mask by removing reflected object parts.

AB - Extraction of foreground is a basic task in surveillance video analysis. In most real cases, its performance is heavily based on the efficiency of shadow detection and on the analysis of lighting conditions and reflections caused by mirrors or other reflective surfaces. This correspondence is focused on the improvement of foreground extraction in the case of planar reflective surfaces. We show that the geometric model of a scene with a planar reflective surface is reduced to the estimation of vanishing-point for the case of an auto-epipolar (skew-symmetric) fundamental matrix. The correspondences for the vanishing-point estimation are extracted from motion statistics. The knowledge of the position of the vanishing point allows us to integrate the geometric model and the motion statistics into image foreground-extraction to separate foreground from reflections, and thus to achieve better performance. The experiments confirm the accuracy of the vanishing point and the improvement of the foreground image mask by removing reflected object parts.

KW - Image and video processing

KW - Robust estimation

KW - Scene analysis

UR - http://www.scopus.com/inward/record.url?scp=67349276359&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67349276359&partnerID=8YFLogxK

U2 - 10.1109/TIP.2009.2017137

DO - 10.1109/TIP.2009.2017137

M3 - Article

C2 - 19366642

AN - SCOPUS:67349276359

VL - 18

SP - 1366

EP - 1372

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 6

ER -