Publication detail

Robust motion segmentation for on-line application

KLICNAR, L. BERAN, V.

Original Title

Robust motion segmentation for on-line application

English Title

Robust motion segmentation for on-line application

Type

conference paper

Language

en

Original Abstract

This paper presents a novel approach for on-line video motion segmentation. Common methods were designed for off-line processing, where time to process one frame is not so important and varies from minutes to hours. The motivation of our work was an application in robotic perception, where a high computational speed is required. The main contribution of this work is an adaptation of existing methods to a higher computational speed and on-line processing. The proposed approach is based on sparse features, we utilized the KLT tracker to obtain their trajectories. A RANSAC-based method is used for initial motion segmentation, resulting motion groups are partitioned by a spatial-proximity constraints. The correspondence of motion groups across frames is solved by one-frame label propagation in forward and backward directions. Finally, an approximation of dense image segmentation is obtained by using the Voronoi tessellation.

English abstract

This paper presents a novel approach for on-line video motion segmentation. Common methods were designed for off-line processing, where time to process one frame is not so important and varies from minutes to hours. The motivation of our work was an application in robotic perception, where a high computational speed is required. The main contribution of this work is an adaptation of existing methods to a higher computational speed and on-line processing. The proposed approach is based on sparse features, we utilized the KLT tracker to obtain their trajectories. A RANSAC-based method is used for initial motion segmentation, resulting motion groups are partitioned by a spatial-proximity constraints. The correspondence of motion groups across frames is solved by one-frame label propagation in forward and backward directions. Finally, an approximation of dense image segmentation is obtained by using the Voronoi tessellation.

Keywords

Motion segmentation, moving objects detection, KLT tracker, Voronoi tessellation.

RIV year

2012

Released

01.12.2012

Publisher

University of West Bohemia in Pilsen

Location

Plzeň

ISBN

978-80-86943-79-4

Book

Proceedings of WSCG'12

Edition

20th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2012 - Conference Proceedings

Edition number

NEUVEDEN

Pages from

205

Pages to

212

Pages count

7

URL

Documents

BibTex


@inproceedings{BUT91497,
  author="Lukáš {Klicnar} and Vítězslav {Beran}",
  title="Robust motion segmentation for on-line application",
  annote="This paper presents a novel approach for on-line video motion segmentation.
Common methods were designed for off-line processing, where time to process one
frame is not so important and varies from minutes to hours. The motivation of our
work was an application in robotic perception, where a high computational speed
is required. The main contribution of this work is an adaptation of existing
methods to a higher computational speed and on-line processing. The proposed
approach is based on sparse features, we utilized the KLT tracker to obtain their
trajectories. A RANSAC-based method is used for initial motion segmentation,
resulting motion groups are partitioned by a spatial-proximity constraints. The
correspondence of motion groups across frames is solved by one-frame label
propagation in forward and backward directions. Finally, an approximation of
dense image segmentation is obtained by using the Voronoi tessellation.",
  address="University of West Bohemia in Pilsen",
  booktitle="Proceedings of WSCG'12",
  chapter="91497",
  edition="20th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2012 - Conference Proceedings",
  howpublished="online",
  institution="University of West Bohemia in Pilsen",
  year="2012",
  month="december",
  pages="205--212",
  publisher="University of West Bohemia in Pilsen",
  type="conference paper"
}