Localizing people in multi-view environment using height map reconstruction in real-time

Ákos Kiss, T. Szirányi

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this article we address the problem of visual people localization, based on the detection of their feet. Localization is based on searching cone intersections. The altitude of location is also retrieved, which eliminates the need of planar ground - which is a common restriction in the related literature. We found that positions can be computed accurately, and despite a large number of false positives, the height map of the scene can be reconstructed with small error. Precision of the detector can be increased given the height map, so that results of our method are comparable to state of the art methods in case of planar ground, but adding the ability to handle arbitrary ground. Our algorithm is capable of real-time operation, based on two optimizations: decreasing the number of cones, and approximating intersection bodies. Cones are back-projections of ellipses in images covering feet regions. Moreover, most demanding steps are parallelizable, and distributable due to lack of data dependencies.

Original languageEnglish
Pages (from-to)2135-2143
Number of pages9
JournalPattern Recognition Letters
Volume34
Issue number16
DOIs
Publication statusPublished - 2013

Fingerprint

Cones
Detectors

Keywords

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

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Localizing people in multi-view environment using height map reconstruction in real-time. / Kiss, Ákos; Szirányi, T.

In: Pattern Recognition Letters, Vol. 34, No. 16, 2013, p. 2135-2143.

Research output: Contribution to journalArticle

@article{68fbb68d8144402d944dad4453768bf0,
title = "Localizing people in multi-view environment using height map reconstruction in real-time",
abstract = "In this article we address the problem of visual people localization, based on the detection of their feet. Localization is based on searching cone intersections. The altitude of location is also retrieved, which eliminates the need of planar ground - which is a common restriction in the related literature. We found that positions can be computed accurately, and despite a large number of false positives, the height map of the scene can be reconstructed with small error. Precision of the detector can be increased given the height map, so that results of our method are comparable to state of the art methods in case of planar ground, but adding the ability to handle arbitrary ground. Our algorithm is capable of real-time operation, based on two optimizations: decreasing the number of cones, and approximating intersection bodies. Cones are back-projections of ellipses in images covering feet regions. Moreover, most demanding steps are parallelizable, and distributable due to lack of data dependencies.",
keywords = "3D position, Multi-view localization, Projection, Real-time processing",
author = "{\'A}kos Kiss and T. Szir{\'a}nyi",
year = "2013",
doi = "10.1016/j.patrec.2013.08.007",
language = "English",
volume = "34",
pages = "2135--2143",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "16",

}

TY - JOUR

T1 - Localizing people in multi-view environment using height map reconstruction in real-time

AU - Kiss, Ákos

AU - Szirányi, T.

PY - 2013

Y1 - 2013

N2 - In this article we address the problem of visual people localization, based on the detection of their feet. Localization is based on searching cone intersections. The altitude of location is also retrieved, which eliminates the need of planar ground - which is a common restriction in the related literature. We found that positions can be computed accurately, and despite a large number of false positives, the height map of the scene can be reconstructed with small error. Precision of the detector can be increased given the height map, so that results of our method are comparable to state of the art methods in case of planar ground, but adding the ability to handle arbitrary ground. Our algorithm is capable of real-time operation, based on two optimizations: decreasing the number of cones, and approximating intersection bodies. Cones are back-projections of ellipses in images covering feet regions. Moreover, most demanding steps are parallelizable, and distributable due to lack of data dependencies.

AB - In this article we address the problem of visual people localization, based on the detection of their feet. Localization is based on searching cone intersections. The altitude of location is also retrieved, which eliminates the need of planar ground - which is a common restriction in the related literature. We found that positions can be computed accurately, and despite a large number of false positives, the height map of the scene can be reconstructed with small error. Precision of the detector can be increased given the height map, so that results of our method are comparable to state of the art methods in case of planar ground, but adding the ability to handle arbitrary ground. Our algorithm is capable of real-time operation, based on two optimizations: decreasing the number of cones, and approximating intersection bodies. Cones are back-projections of ellipses in images covering feet regions. Moreover, most demanding steps are parallelizable, and distributable due to lack of data dependencies.

KW - 3D position

KW - Multi-view localization

KW - Projection

KW - Real-time processing

UR - http://www.scopus.com/inward/record.url?scp=84883891843&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883891843&partnerID=8YFLogxK

U2 - 10.1016/j.patrec.2013.08.007

DO - 10.1016/j.patrec.2013.08.007

M3 - Article

VL - 34

SP - 2135

EP - 2143

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 16

ER -