Tracking interacting targets with laser scanner via on-line supervised learning

Successful multi-target tracking requires locating the targets and labeling their identities. For the laser based tracking system, the latter becomes significantly more challenging when the targets frequently interact with each other. This paper presents a novel on-line supervised learning based met...

Full description

Saved in:
Bibliographic Details
Published in2008 IEEE International Conference on Robotics and Automation pp. 2271 - 2276
Main Authors Xuan Song, Jinshi Cui, Xulei Wang, Huijing Zhao, Hongbin Zha
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2008
Subjects
Computational complexity
Filters
Humans
Laboratories
Legged locomotion
Merging
Robotics and automation
Supervised learning
Target tracking
USA Councils
Online AccessGet full text

Cover

More Information
Summary:Successful multi-target tracking requires locating the targets and labeling their identities. For the laser based tracking system, the latter becomes significantly more challenging when the targets frequently interact with each other. This paper presents a novel on-line supervised learning based method for tracking interacting targets with laser scanner. When the targets do not interact with each other, we collect samples and train a classifier for each target. When the targets are in close proximity, we use these classifiers to assist in tracking. Different evaluations demonstrate that this method has a better tracking performance than previous methods when interactions occur, and can maintain correct tracking under various complex tracking situations.
ISBN:1424416469
9781424416462
ISSN:1050-4729
2577-087X
DOI:10.1109/ROBOT.2008.4543552