Home / Issues / 11.2014 / Learning Two-Person Interaction Models for Responsive Synthetic Humanoids
Document Actions

Citation and metadata

Back to the article.

Recommended citation

David Vogt, Heni Ben Amor, Erik Berger, and Bernhard Jung, Learning Two-Person Interaction Models for Responsive Synthetic Humanoids. Journal of Virtual Reality and Broadcastings, 11(2014), no. 1. (urn:nbn:de:0009-6-38565)

Download Citation

Endnote

%0 Journal Article
%T Learning Two-Person Interaction Models for Responsive Synthetic Humanoids
%A Vogt, David
%A Ben Amor, Heni
%A Berger, Erik
%A Jung, Bernhard
%J Journal of Virtual Reality and Broadcastings
%D 2014
%V 11(2014)
%N 1
%@ 1860-2037
%F vogt2014
%X Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.
%L 004
%K humanoid robots
%K imitation learning
%K interaction learning
%K motion adaptation
%K motor learning
%K virtual characters
%R 10.20385/1860-2037/11.2014.1
%U http://nbn-resolving.de/urn:nbn:de:0009-6-38565
%U http://dx.doi.org/10.20385/1860-2037/11.2014.1
Download

Bibtex

@Article{vogt2014,
  author = 	"Vogt, David
		and Ben Amor, Heni
		and Berger, Erik
		and Jung, Bernhard",
  title = 	"Learning Two-Person Interaction Models for Responsive Synthetic Humanoids",
  journal = 	"Journal of Virtual Reality and Broadcastings",
  year = 	"2014",
  volume = 	"11(2014)",
  number = 	"1",
  keywords = 	"humanoid robots; imitation learning; interaction learning; motion adaptation; motor learning; virtual characters",
  abstract = 	"Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.",
  issn = 	"1860-2037",
  doi = 	"10.20385/1860-2037/11.2014.1",
  url = 	"http://nbn-resolving.de/urn:nbn:de:0009-6-38565"
}
Download

RIS

TY  - JOUR
AU  - Vogt, David
AU  - Ben Amor, Heni
AU  - Berger, Erik
AU  - Jung, Bernhard
PY  - 2014
DA  - 2014//
TI  - Learning Two-Person Interaction Models for Responsive Synthetic Humanoids
JO  - Journal of Virtual Reality and Broadcastings
VL  - 11(2014)
IS  - 1
KW  - humanoid robots
KW  - imitation learning
KW  - interaction learning
KW  - motion adaptation
KW  - motor learning
KW  - virtual characters
AB  - Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.
SN  - 1860-2037
UR  - http://nbn-resolving.de/urn:nbn:de:0009-6-38565
DO  - 10.20385/1860-2037/11.2014.1
ID  - vogt2014
ER  -
Download

Wordbib

<?xml version="1.0" encoding="UTF-8"?>
<b:Sources SelectedStyle="" xmlns:b="http://schemas.openxmlformats.org/officeDocument/2006/bibliography"  xmlns="http://schemas.openxmlformats.org/officeDocument/2006/bibliography" >
<b:Source>
<b:Tag>vogt2014</b:Tag>
<b:SourceType>ArticleInAPeriodical</b:SourceType>
<b:Year>2014</b:Year>
<b:PeriodicalTitle>Journal of Virtual Reality and Broadcastings</b:PeriodicalTitle>
<b:Volume>11(2014)</b:Volume>
<b:Issue>1</b:Issue>
<b:Url>http://nbn-resolving.de/urn:nbn:de:0009-6-38565</b:Url>
<b:Url>http://dx.doi.org/10.20385/1860-2037/11.2014.1</b:Url>
<b:Author>
<b:Author><b:NameList>
<b:Person><b:Last>Vogt</b:Last><b:First>David</b:First></b:Person>
<b:Person><b:Last>Ben Amor</b:Last><b:First>Heni</b:First></b:Person>
<b:Person><b:Last>Berger</b:Last><b:First>Erik</b:First></b:Person>
<b:Person><b:Last>Jung</b:Last><b:First>Bernhard</b:First></b:Person>
</b:NameList></b:Author>
</b:Author>
<b:Title>Learning Two-Person Interaction Models for Responsive Synthetic Humanoids</b:Title>
<b:Comments>Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.</b:Comments>
</b:Source>
</b:Sources>
Download

ISI

PT Journal
AU Vogt, D
   Ben Amor, H
   Berger, E
   Jung, B
TI Learning Two-Person Interaction Models for Responsive Synthetic Humanoids
SO Journal of Virtual Reality and Broadcastings
PY 2014
VL 11(2014)
IS 1
DI 10.20385/1860-2037/11.2014.1
DE humanoid robots; imitation learning; interaction learning; motion adaptation; motor learning; virtual characters
AB Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.
ER
Download

Mods

<mods>
  <titleInfo>
    <title>Learning Two-Person Interaction Models for Responsive Synthetic Humanoids</title>
  </titleInfo>
  <name type="personal">
    <namePart type="family">Vogt</namePart>
    <namePart type="given">David</namePart>
  </name>
  <name type="personal">
    <namePart type="family">Ben Amor</namePart>
    <namePart type="given">Heni</namePart>
  </name>
  <name type="personal">
    <namePart type="family">Berger</namePart>
    <namePart type="given">Erik</namePart>
  </name>
  <name type="personal">
    <namePart type="family">Jung</namePart>
    <namePart type="given">Bernhard</namePart>
  </name>
  <abstract>Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.</abstract>
  <subject>
    <topic>humanoid robots</topic>
    <topic>imitation learning</topic>
    <topic>interaction learning</topic>
    <topic>motion adaptation</topic>
    <topic>motor learning</topic>
    <topic>virtual characters</topic>
  </subject>
  <classification authority="ddc">004</classification>
  <relatedItem type="host">
    <genre authority="marcgt">periodical</genre>
    <genre>academic journal</genre>
    <titleInfo>
      <title>Journal of Virtual Reality and Broadcastings</title>
    </titleInfo>
    <part>
      <detail type="volume">
        <number>11(2014)</number>
      </detail>
      <detail type="issue">
        <number>1</number>
      </detail>
      <date>2014</date>
    </part>
  </relatedItem>
  <identifier type="issn">1860-2037</identifier>
  <identifier type="urn">urn:nbn:de:0009-6-38565</identifier>
  <identifier type="doi">10.20385/1860-2037/11.2014.1</identifier>
  <identifier type="uri">http://nbn-resolving.de/urn:nbn:de:0009-6-38565</identifier>
  <identifier type="citekey">vogt2014</identifier>
</mods>
Download

Full Metadata

Bibliographic Citation JVRB, 11(2014), no. 1.
Title

Learning Two-Person Interaction Models for Responsive Synthetic Humanoids (eng)
Author David Vogt, Heni Ben Amor, Erik Berger, Bernhard Jung
Language eng
Abstract Imitation learning is a promising approach for generating life-like behaviors of virtual humans and humanoid robots. So far, however, imitation learning has been mostly restricted to single agent settings where observed motions are adapted to new environment conditions but not to the dynamic behavior of interaction partners. In this paper, we introduce a new imitation learning approach that is based on the simultaneous motion capture of two human interaction partners. From the observed interactions, low-dimensional motion models are extracted and a mapping between these motion models is learned. This interaction model allows the real-time generation of agent behaviors that are responsive to the body movements of an interaction partner. The interaction model can be applied both to the animation of virtual characters as well as to the behavior generation for humanoid robots.
Subject humanoid robots, imitation learning, interaction learning, motion adaptation, motor learning, virtual characters
Classified Subjects
swd
  • 4170603-1, Motorisches Lernen
  • 7576811-2, Humanoider Roboter
DDC 004
Rights DPPL
URN: urn:nbn:de:0009-6-38565
DOI https://doi.org/10.20385/1860-2037/11.2014.1