Vision based people tracking using stereo camera and camshift

Kang Il Park, Chan Ik Park, Jangmyung Lee

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

In this paper proposes a visual manipulation method that incorporates computation of 3D object position from characteristic disparity of stereo image. First, we proceed to an image revision work using the calibration. To obtain the characteristic disparity, a set of characteristic points, essential to obtain characteristics like center points or corner points are determined from the CAMShift center of left and right images. By iterative computation and tracking in real time, the algorithm is able to robustly provide accurate 3D position of target objects.

Original languageEnglish
Title of host publicationIntelligent Robotics and Applications - 6th International Conference, ICIRA 2013, Proceedings
PublisherSpringer Verlag
Pages145-153
Number of pages9
EditionPART 1
ISBN (Print)9783642408519
DOIs
StatePublished - 2013
Event6th International Conference on Intelligent Robotics and Applications, ICIRA 2013 - Busan, Korea, Republic of
Duration: 25 Sep 201328 Sep 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume8102 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th International Conference on Intelligent Robotics and Applications, ICIRA 2013
Country/TerritoryKorea, Republic of
CityBusan
Period25/09/1328/09/13

Keywords

  • Calibration
  • CAMShift Algorithm
  • People Tracking
  • Stereo Camera

Fingerprint

Dive into the research topics of 'Vision based people tracking using stereo camera and camshift'. Together they form a unique fingerprint.

Cite this