Home / Issues / 7.2010 / Algorithms For Automatic And Robust Registration Of 3D Head Scans
Document Actions

Citation and metadata

Back to the article.

Recommended citation

David C. Schneider, and Peter Eisert, Algorithms For Automatic And Robust Registration Of 3D Head Scans. JVRB - Journal of Virtual Reality and Broadcasting, 7(2010), no. 7. (urn:nbn:de:0009-6-26626)

Download Citation

Endnote

%0 Journal Article
%T Algorithms For Automatic And Robust Registration Of 3D Head Scans
%A Schneider, David C.
%A Eisert, Peter
%J JVRB - Journal of Virtual Reality and Broadcasting
%D 2010
%V 7(2010)
%N 7
%@ 1860-2037
%F schneider2010
%X wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as "shape completion". The importance of regularization in the case of extreme shape completion is shown.
%L 004
%K 3D face processing
%K ICP
%K geometry interpolation
%K morphable head models
%K nonrigid registration
%K shape completion
%R 10.20385/1860-2037/7.2010.7
%U http://nbn-resolving.de/urn:nbn:de:0009-6-26626
%U http://dx.doi.org/10.20385/1860-2037/7.2010.7
Download

Bibtex

@Article{schneider2010,
  author = 	"Schneider, David C.
		and Eisert, Peter",
  title = 	"Algorithms For Automatic And Robust Registration Of 3D Head Scans",
  journal = 	"JVRB - Journal of Virtual Reality and Broadcasting",
  year = 	"2010",
  volume = 	"7(2010)",
  number = 	"7",
  keywords = 	"3D face processing; ICP; geometry interpolation; morphable head models; nonrigid registration; shape completion",
  abstract = 	"wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as ``shape completion''. The importance of regularization in the case of extreme shape completion is shown.",
  issn = 	"1860-2037",
  doi = 	"10.20385/1860-2037/7.2010.7",
  url = 	"http://nbn-resolving.de/urn:nbn:de:0009-6-26626"
}
Download

RIS

TY  - JOUR
AU  - Schneider, David C.
AU  - Eisert, Peter
PY  - 2010
DA  - 2010//
TI  - Algorithms For Automatic And Robust Registration Of 3D Head Scans
JO  - JVRB - Journal of Virtual Reality and Broadcasting
VL  - 7(2010)
IS  - 7
KW  - 3D face processing
KW  - ICP
KW  - geometry interpolation
KW  - morphable head models
KW  - nonrigid registration
KW  - shape completion
AB  - wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as "shape completion". The importance of regularization in the case of extreme shape completion is shown.
SN  - 1860-2037
UR  - http://nbn-resolving.de/urn:nbn:de:0009-6-26626
DO  - 10.20385/1860-2037/7.2010.7
ID  - schneider2010
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>schneider2010</b:Tag>
<b:SourceType>ArticleInAPeriodical</b:SourceType>
<b:Year>2010</b:Year>
<b:PeriodicalTitle>JVRB - Journal of Virtual Reality and Broadcasting</b:PeriodicalTitle>
<b:Volume>7(2010)</b:Volume>
<b:Issue>7</b:Issue>
<b:Url>http://nbn-resolving.de/urn:nbn:de:0009-6-26626</b:Url>
<b:Url>http://dx.doi.org/10.20385/1860-2037/7.2010.7</b:Url>
<b:Author>
<b:Author><b:NameList>
<b:Person><b:Last>Schneider</b:Last><b:First>David C.</b:First></b:Person>
<b:Person><b:Last>Eisert</b:Last><b:First>Peter</b:First></b:Person>
</b:NameList></b:Author>
</b:Author>
<b:Title>Algorithms For Automatic And Robust Registration Of 3D Head Scans</b:Title>
<b:Comments>wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as &quot;shape completion&quot;. The importance of regularization in the case of extreme shape completion is shown.</b:Comments>
</b:Source>
</b:Sources>
Download

ISI

PT Journal
AU Schneider, D
   Eisert, P
TI Algorithms For Automatic And Robust Registration Of 3D Head Scans
SO JVRB - Journal of Virtual Reality and Broadcasting
PY 2010
VL 7(2010)
IS 7
DI 10.20385/1860-2037/7.2010.7
DE 3D face processing; ICP; geometry interpolation; morphable head models; nonrigid registration; shape completion
AB wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as "shape completion". The importance of regularization in the case of extreme shape completion is shown.
ER
Download

Mods

<mods>
  <titleInfo>
    <title>Algorithms For Automatic And Robust Registration Of 3D Head Scans</title>
  </titleInfo>
  <name type="personal">
    <namePart type="family">Schneider</namePart>
    <namePart type="given">David C.</namePart>
  </name>
  <name type="personal">
    <namePart type="family">Eisert</namePart>
    <namePart type="given">Peter</namePart>
  </name>
  <abstract>wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as "shape completion". The importance of regularization in the case of extreme shape completion is shown.</abstract>
  <subject>
    <topic>3D face processing</topic>
    <topic>ICP</topic>
    <topic>geometry interpolation</topic>
    <topic>morphable head models</topic>
    <topic>nonrigid registration</topic>
    <topic>shape completion</topic>
  </subject>
  <classification authority="ddc">004</classification>
  <relatedItem type="host">
    <genre authority="marcgt">periodical</genre>
    <genre>academic journal</genre>
    <titleInfo>
      <title>JVRB - Journal of Virtual Reality and Broadcasting</title>
    </titleInfo>
    <part>
      <detail type="volume">
        <number>7(2010)</number>
      </detail>
      <detail type="issue">
        <number>7</number>
      </detail>
      <date>2010</date>
    </part>
  </relatedItem>
  <identifier type="issn">1860-2037</identifier>
  <identifier type="urn">urn:nbn:de:0009-6-26626</identifier>
  <identifier type="doi">10.20385/1860-2037/7.2010.7</identifier>
  <identifier type="uri">http://nbn-resolving.de/urn:nbn:de:0009-6-26626</identifier>
  <identifier type="citekey">schneider2010</identifier>
</mods>
Download

Full Metadata

Bibliographic Citation JVRB, 7(2010), no. 7.
Title

Algorithms For Automatic And Robust Registration Of 3D Head Scans (eng)
Author David C. Schneider, Peter Eisert
Language eng
Abstract wo methods for registering laser-scans of human heads and transforming them to a new semantically consistent topology defined by a user-provided template mesh are described. Both algorithms are stated within the Iterative Closest Point framework. The first method is based on finding landmark correspondences by iteratively registering the vicinity of a landmark with a re-weighted error function. Thin-plate spline interpolation is then used to deform the template mesh and finally the scan is resampled in the topology of the deformed template. The second algorithm employs a morphable shape model, which can be computed from a database of laser-scans using the first algorithm. It directly optimizes pose and shape of the morphable model. The use of the algorithm with PCA mixture models, where the shape is split up into regions each described by an individual subspace, is addressed. Mixture models require either blending or regularization strategies, both of which are described in detail. For both algorithms, strategies for filling in missing geometry for incomplete laser-scans are described. While an interpolation-based approach can be used to fill in small or smooth regions, the model-driven algorithm is capable of fitting a plausible complete head mesh to arbitrarily small geometry, which is known as "shape completion". The importance of regularization in the case of extreme shape completion is shown.
Subject 3D face processing, ICP, geometry interpolation, morphable head models, nonrigid registration, shape completion
Classified Subjects
swd
  • 7704418-6, 3D-Scanner
  • 4130267-9, Computational geometry
  • 4128931-6, Kopf
DDC 004
Rights DPPL
URN: urn:nbn:de:0009-6-26626
DOI https://doi.org/10.20385/1860-2037/7.2010.7