Publication detail

Depth-Based Filtration for Tracking Boost

CHRÁPEK, D. BERAN, V. ZEMČÍK, P.

Original Title

Depth-Based Filtration for Tracking Boost

English Title

Depth-Based Filtration for Tracking Boost

Type

conference paper

Language

en

Original Abstract

This paper presents a novel depth information utilization method for performance boosting of tracking in traditional RGB trackers for arbitrary objects (objects not known in advance) by object segmentation/separation supported by depth information. The main focus is on real-time applications, such as robotics or surveillance, where exploitation of depth sensors, that are nowadays affordable, is not only possible but also feasible. The aim is to show that the depth information used for target segmentation significantly helps reducing incorrect model updates caused by occlusion or drifts and improves success rate and precision of traditional RGB tracker while keeping comparably efficient and thus possibly real-time. The paper also presents and discusses the achieved performance results.

English abstract

This paper presents a novel depth information utilization method for performance boosting of tracking in traditional RGB trackers for arbitrary objects (objects not known in advance) by object segmentation/separation supported by depth information. The main focus is on real-time applications, such as robotics or surveillance, where exploitation of depth sensors, that are nowadays affordable, is not only possible but also feasible. The aim is to show that the depth information used for target segmentation significantly helps reducing incorrect model updates caused by occlusion or drifts and improves success rate and precision of traditional RGB tracker while keeping comparably efficient and thus possibly real-time. The paper also presents and discusses the achieved performance results.

Keywords

Real-time, RGBD, Segmentation, Tracking

RIV year

2015

Released

06.11.2015

Publisher

Springer International Publishing

Location

Catania

ISBN

978-3-319-25903-1

Book

Springer International Publishing

Edition

Lecture Notes in Computer Science

Edition number

NEUVEDEN

Pages from

217

Pages to

228

Pages count

12

URL

Documents

BibTex


@inproceedings{BUT119922,
  author="David {Chrápek} and Vítězslav {Beran} and Pavel {Zemčík}",
  title="Depth-Based Filtration for Tracking Boost",
  annote="This paper presents a novel depth information utilization method for performance
boosting of tracking in traditional RGB trackers for arbitrary objects (objects
not known in advance) by object segmentation/separation supported by depth
information. The main focus is on real-time applications, such as robotics or
surveillance, where exploitation of depth sensors, that are nowadays affordable,
is not only possible but also feasible. The aim is to show that the depth
information used for target segmentation significantly helps reducing incorrect
model updates caused by occlusion or drifts and improves success rate and
precision of traditional RGB tracker while keeping comparably efficient and thus
possibly real-time. The paper also presents and discusses the achieved
performance results.",
  address="Springer International Publishing",
  booktitle="Springer International Publishing",
  chapter="119922",
  doi="10.1007/978-3-319-25903-1_19",
  edition="Lecture Notes in Computer Science",
  howpublished="online",
  institution="Springer International Publishing",
  number="9386",
  year="2015",
  month="november",
  pages="217--228",
  publisher="Springer International Publishing",
  type="conference paper"
}