Multi-view people detection on arbitrary ground in real-time

Ákos Kiss, T. Szirányi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

We show a method to detect accurate 3D position of people from multiple views, regardless of the geometry of the ground. In our new method we search for intersections of 3D primitives (cones) to find positions of feet. The cones are computed by back-projecting ellipses covering feet in input images. Instead of computing complex intersection body, we use approximation to speed up intersection computing. We found that feet positions are determined accurately, and the height map of the ground can be reconstructed with small error. We compared our method to other multiview-detectors - using somewhat different test methodology -, and achieved comparable results, with the benefit of handling arbitrary ground. We also present accurately reconstructed height map of non-planar ground. Our algorithm is fast and most of steps are parallelizable, making it possibly available for smart camera systems.

Original languageEnglish
Title of host publicationVISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications
Pages675-680
Number of pages6
Volume1
Publication statusPublished - 2013
Event8th International Conference on Computer Vision Theory and Applications, VISAPP 2013 - Barcelona, Spain
Duration: Feb 21 2013Feb 24 2013

Other

Other8th International Conference on Computer Vision Theory and Applications, VISAPP 2013
CountrySpain
CityBarcelona
Period2/21/132/24/13

Fingerprint

Cones
Cameras
Detectors
Geometry

Keywords

  • 3D position
  • Multi-view detection
  • Projection
  • Real-time processing

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Kiss, Á., & Szirányi, T. (2013). Multi-view people detection on arbitrary ground in real-time. In VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications (Vol. 1, pp. 675-680)

Multi-view people detection on arbitrary ground in real-time. / Kiss, Ákos; Szirányi, T.

VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1 2013. p. 675-680.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kiss, Á & Szirányi, T 2013, Multi-view people detection on arbitrary ground in real-time. in VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. vol. 1, pp. 675-680, 8th International Conference on Computer Vision Theory and Applications, VISAPP 2013, Barcelona, Spain, 2/21/13.
Kiss Á, Szirányi T. Multi-view people detection on arbitrary ground in real-time. In VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1. 2013. p. 675-680
Kiss, Ákos ; Szirányi, T. / Multi-view people detection on arbitrary ground in real-time. VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1 2013. pp. 675-680
@inproceedings{da6e0cdb286140d18e9e701497a021f6,
title = "Multi-view people detection on arbitrary ground in real-time",
abstract = "We show a method to detect accurate 3D position of people from multiple views, regardless of the geometry of the ground. In our new method we search for intersections of 3D primitives (cones) to find positions of feet. The cones are computed by back-projecting ellipses covering feet in input images. Instead of computing complex intersection body, we use approximation to speed up intersection computing. We found that feet positions are determined accurately, and the height map of the ground can be reconstructed with small error. We compared our method to other multiview-detectors - using somewhat different test methodology -, and achieved comparable results, with the benefit of handling arbitrary ground. We also present accurately reconstructed height map of non-planar ground. Our algorithm is fast and most of steps are parallelizable, making it possibly available for smart camera systems.",
keywords = "3D position, Multi-view detection, Projection, Real-time processing",
author = "{\'A}kos Kiss and T. Szir{\'a}nyi",
year = "2013",
language = "English",
isbn = "9789898565471",
volume = "1",
pages = "675--680",
booktitle = "VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications",

}

TY - GEN

T1 - Multi-view people detection on arbitrary ground in real-time

AU - Kiss, Ákos

AU - Szirányi, T.

PY - 2013

Y1 - 2013

N2 - We show a method to detect accurate 3D position of people from multiple views, regardless of the geometry of the ground. In our new method we search for intersections of 3D primitives (cones) to find positions of feet. The cones are computed by back-projecting ellipses covering feet in input images. Instead of computing complex intersection body, we use approximation to speed up intersection computing. We found that feet positions are determined accurately, and the height map of the ground can be reconstructed with small error. We compared our method to other multiview-detectors - using somewhat different test methodology -, and achieved comparable results, with the benefit of handling arbitrary ground. We also present accurately reconstructed height map of non-planar ground. Our algorithm is fast and most of steps are parallelizable, making it possibly available for smart camera systems.

AB - We show a method to detect accurate 3D position of people from multiple views, regardless of the geometry of the ground. In our new method we search for intersections of 3D primitives (cones) to find positions of feet. The cones are computed by back-projecting ellipses covering feet in input images. Instead of computing complex intersection body, we use approximation to speed up intersection computing. We found that feet positions are determined accurately, and the height map of the ground can be reconstructed with small error. We compared our method to other multiview-detectors - using somewhat different test methodology -, and achieved comparable results, with the benefit of handling arbitrary ground. We also present accurately reconstructed height map of non-planar ground. Our algorithm is fast and most of steps are parallelizable, making it possibly available for smart camera systems.

KW - 3D position

KW - Multi-view detection

KW - Projection

KW - Real-time processing

UR - http://www.scopus.com/inward/record.url?scp=84878254663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878254663&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84878254663

SN - 9789898565471

VL - 1

SP - 675

EP - 680

BT - VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications

ER -