Background Subtraction in Video Using Recursive Mixture Models, Spatio-Temporal Filtering and Shadow Removal

Zezhi Chen; Nick Pears; Michael Freeman; Jim Austin

doi:10.1007/978-3-642-10520-3_109

Background Subtraction in Video Using Recursive Mixture Models, Spatio-Temporal Filtering and Shadow Removal

Zezhi Chen, Nick Pears, Michael Freeman, Jim Austin

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

We describe our approach to segmenting moving objects from the color video data supplied by a nominally stationary camera. There are two main contributions in our work. The first contribution augments Zivkovic and Heijden’s recursively updated Gaussian mixture model approach, with a multi-dimensional Gaussian kernel spatio-temporal smoothing transform. We show that this improves the segmentation performance of the original approach, particularly in adverse imaging conditions, such as when there is camera vibration. Our second contribution is to present a comprehensive comparative evaluation of shadow and highlight detection appoaches, which is an essential component of background subtraction in unconstrained outdoor scenes. A comparative evelaution of these approaches over different color-spaces is currently lacking in the literature. We show that both segmentation and shadow removal performs best when we use RGB color spaces.

Original language	Undefined/Unknown
Pages	1141-1150
DOIs	https://doi.org/10.1007/978-3-642-10520-3_109
Publication status	Published - 2009

Access to Document

10.1007/978-3-642-10520-3_109

http://dx.doi.org/10.1007/978-3-642-10520-3_109

Cite this

@conference{6108cc7a4d214f5ea7c5d070c81eb2ac,

title = "Background Subtraction in Video Using Recursive Mixture Models, Spatio-Temporal Filtering and Shadow Removal",

abstract = "We describe our approach to segmenting moving objects from the color video data supplied by a nominally stationary camera. There are two main contributions in our work. The first contribution augments Zivkovic and Heijden{\textquoteright}s recursively updated Gaussian mixture model approach, with a multi-dimensional Gaussian kernel spatio-temporal smoothing transform. We show that this improves the segmentation performance of the original approach, particularly in adverse imaging conditions, such as when there is camera vibration. Our second contribution is to present a comprehensive comparative evaluation of shadow and highlight detection appoaches, which is an essential component of background subtraction in unconstrained outdoor scenes. A comparative evelaution of these approaches over different color-spaces is currently lacking in the literature. We show that both segmentation and shadow removal performs best when we use RGB color spaces. ",

author = "Zezhi Chen and Nick Pears and Michael Freeman and Jim Austin",

year = "2009",

doi = "10.1007/978-3-642-10520-3_109",

language = "Undefined/Unknown",

pages = "1141--1150",

}

TY - CONF

T1 - Background Subtraction in Video Using Recursive Mixture Models, Spatio-Temporal Filtering and Shadow Removal

AU - Chen, Zezhi

AU - Pears, Nick

AU - Freeman, Michael

AU - Austin, Jim

PY - 2009

Y1 - 2009

N2 - We describe our approach to segmenting moving objects from the color video data supplied by a nominally stationary camera. There are two main contributions in our work. The first contribution augments Zivkovic and Heijden’s recursively updated Gaussian mixture model approach, with a multi-dimensional Gaussian kernel spatio-temporal smoothing transform. We show that this improves the segmentation performance of the original approach, particularly in adverse imaging conditions, such as when there is camera vibration. Our second contribution is to present a comprehensive comparative evaluation of shadow and highlight detection appoaches, which is an essential component of background subtraction in unconstrained outdoor scenes. A comparative evelaution of these approaches over different color-spaces is currently lacking in the literature. We show that both segmentation and shadow removal performs best when we use RGB color spaces.

AB - We describe our approach to segmenting moving objects from the color video data supplied by a nominally stationary camera. There are two main contributions in our work. The first contribution augments Zivkovic and Heijden’s recursively updated Gaussian mixture model approach, with a multi-dimensional Gaussian kernel spatio-temporal smoothing transform. We show that this improves the segmentation performance of the original approach, particularly in adverse imaging conditions, such as when there is camera vibration. Our second contribution is to present a comprehensive comparative evaluation of shadow and highlight detection appoaches, which is an essential component of background subtraction in unconstrained outdoor scenes. A comparative evelaution of these approaches over different color-spaces is currently lacking in the literature. We show that both segmentation and shadow removal performs best when we use RGB color spaces.

UR - http://www.scopus.com/inward/record.url?scp=72449179513&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-10520-3_109

DO - 10.1007/978-3-642-10520-3_109

M3 - Paper

SP - 1141

EP - 1150

ER -

Background Subtraction in Video Using Recursive Mixture Models, Spatio-Temporal Filtering and Shadow Removal

Abstract

Access to Document

Other files and links

Cite this