Self-organizing maps for hand and full body tracking

Foti Coleca; Andreea State; Sascha Klement; Erhardt Barth; Thomas Martinetz

doi:10.1016/j.neucom.2013.10.041

Self-organizing maps for hand and full body tracking

Foti Coleca^*, Andreea State, Sascha Klement, Erhardt Barth, Thomas Martinetz

^*Corresponding author for this work

Institute of Neuro- and Bioinformatics

10 Citations (Scopus)

Abstract

Touch-free gesture technology opens new avenues for human-machine interaction. We show how self-organizing maps (SOM) can be used for hand and full body tracking. We use a range camera for data acquisition and apply a SOM-learning process for each frame in order to capture the pose. In a next step we introduce an extension of the SOM to 1D and 2D segments for an improved representation and skeleton tracking of body and hand. The proposed SOM based algorithms are very efficient and robust, and produce good tracking results. Their efficiency allows to implement these algorithms on embedded systems, which we demonstrate on an ARM-based embedded platform.

Original language	English
Journal	Neurocomputing
Volume	147
Issue number	1
Pages (from-to)	174-184
Number of pages	11
ISSN	0925-2312
DOIs	https://doi.org/10.1016/j.neucom.2013.10.041
Publication status	Published - 01.01.2015
Event	9th Workshop on Self-Organizing Maps - Santiago, Santiago, Chile Duration: 12.12.2012 → 14.12.2012 http://www.die.uchile.cl/wsom2012/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.neucom.2013.10.041

http://www.inb.uni-luebeck.de/publications/pdfs/CoStKlBaMa15.pdf

Cite this

@article{8ab612a7e83344f78aaaaefacb45ee40,

title = "Self-organizing maps for hand and full body tracking",

abstract = "Touch-free gesture technology opens new avenues for human-machine interaction. We show how self-organizing maps (SOM) can be used for hand and full body tracking. We use a range camera for data acquisition and apply a SOM-learning process for each frame in order to capture the pose. In a next step we introduce an extension of the SOM to 1D and 2D segments for an improved representation and skeleton tracking of body and hand. The proposed SOM based algorithms are very efficient and robust, and produce good tracking results. Their efficiency allows to implement these algorithms on embedded systems, which we demonstrate on an ARM-based embedded platform.",

author = "Foti Coleca and Andreea State and Sascha Klement and Erhardt Barth and Thomas Martinetz",

year = "2015",

month = jan,

day = "1",

doi = "10.1016/j.neucom.2013.10.041",

language = "English",

volume = "147",

pages = "174--184",

journal = "Neurocomputing",

issn = "0925-2312",

number = "1",

note = "9th Workshop on Self-Organizing Maps ; Conference date: 12-12-2012 Through 14-12-2012",

url = "http://www.die.uchile.cl/wsom2012/",

}

TY - JOUR

T1 - Self-organizing maps for hand and full body tracking

AU - Coleca, Foti

AU - State, Andreea

AU - Klement, Sascha

AU - Barth, Erhardt

AU - Martinetz, Thomas

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Touch-free gesture technology opens new avenues for human-machine interaction. We show how self-organizing maps (SOM) can be used for hand and full body tracking. We use a range camera for data acquisition and apply a SOM-learning process for each frame in order to capture the pose. In a next step we introduce an extension of the SOM to 1D and 2D segments for an improved representation and skeleton tracking of body and hand. The proposed SOM based algorithms are very efficient and robust, and produce good tracking results. Their efficiency allows to implement these algorithms on embedded systems, which we demonstrate on an ARM-based embedded platform.

AB - Touch-free gesture technology opens new avenues for human-machine interaction. We show how self-organizing maps (SOM) can be used for hand and full body tracking. We use a range camera for data acquisition and apply a SOM-learning process for each frame in order to capture the pose. In a next step we introduce an extension of the SOM to 1D and 2D segments for an improved representation and skeleton tracking of body and hand. The proposed SOM based algorithms are very efficient and robust, and produce good tracking results. Their efficiency allows to implement these algorithms on embedded systems, which we demonstrate on an ARM-based embedded platform.

UR - http://www.scopus.com/inward/record.url?scp=84924039164&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2013.10.041

DO - 10.1016/j.neucom.2013.10.041

M3 - Journal articles

AN - SCOPUS:84924039164

SN - 0925-2312

VL - 147

SP - 174

EP - 184

JO - Neurocomputing

JF - Neurocomputing

IS - 1

T2 - 9th Workshop on Self-Organizing Maps

Y2 - 12 December 2012 through 14 December 2012

ER -

Self-organizing maps for hand and full body tracking

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this