File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Article: Adaptive human motion analysis and prediction
  • Basic View
  • Metadata View
  • XML View
TitleAdaptive human motion analysis and prediction
 
AuthorsChen, Z1
Wang, L1
Yung, NHC1
 
KeywordsAuto regressive models
Human motion analysis
Human motions
Long-term future
Long-term prediction
 
Issue Date2011
 
PublisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/pr
 
CitationPattern Recognition, 2011, v. 44 n. 12, p. 2902-2914 [How to Cite?]
DOI: http://dx.doi.org/10.1016/j.patcog.2011.04.022
 
AbstractHuman motion analysis and prediction is an active research area where predicting human motion is often performed for a single time step based on historical motion. In recent years, longer term human motion prediction has been attempted over a number of future time steps. Most current methods learn motion patterns (MPs) from observed trajectories and then use them for prediction. However, these learned MPs may not be indicative due to inadequate observation, which naturally affects the reliability of motion prediction. In this paper, we present an adaptive human motion analysis and prediction method. It adaptively predicts motion based on the classified MPs in terms of their credibility, which refers to how indicative the learned MPs are for the specific environment. The main contributions of the proposed method are as follows: First, it provides a comprehensive description of MPs including not only the learned MPs but also their evaluated credibility. Second, it predicts long-term future motion with reasonable accuracy. A number of experiments have been conducted in simulated scenes and real-world scenes and the prediction results have been quantitatively evaluated. The results show that the proposed method is effective and superior in its performance when compared with a recursively applied Auto-Regressive (AR) model, which is called the Recursive Short-term Predictor (RSP) for long-term prediction. The proposed method has 17.73% of improvement over the RSP in prediction accuracy in the experiment with the best performance. On average, the proposed method has 5% improvement over the RSP in prediction accuracy over 10 experiments. © 2011 Elsevier Ltd. All rights reserved.
 
ISSN0031-3203
2012 Impact Factor: 2.632
2012 SCImago Journal Rankings: 2.365
 
DOIhttp://dx.doi.org/10.1016/j.patcog.2011.04.022
 
ISI Accession Number IDWOS:000292947000008
Funding AgencyGrant Number
Research Grant Council of the Hong Kong Special Administrative Region, ChinaHKU719406E
HKU719608E
University of Hong Kong
Funding Information:

The work on MP prediction was supported in part by a grant from the Research Grant Council of the Hong Kong Special Administrative Region, China, under Project HKU719406E and in part by the Postgraduate Studentship of The University of Hong Kong. The work on trajectory extraction was supported in part by a grant from the Research Grant Council of the Hong Kong Special Administrative Region, China, under Project HKU719608E and in part by the Postgraduate Studentship of the University of Hong Kong.

 
ReferencesReferences in Scopus
 
DC FieldValue
dc.contributor.authorChen, Z
 
dc.contributor.authorWang, L
 
dc.contributor.authorYung, NHC
 
dc.date.accessioned2012-08-08T08:34:25Z
 
dc.date.available2012-08-08T08:34:25Z
 
dc.date.issued2011
 
dc.description.abstractHuman motion analysis and prediction is an active research area where predicting human motion is often performed for a single time step based on historical motion. In recent years, longer term human motion prediction has been attempted over a number of future time steps. Most current methods learn motion patterns (MPs) from observed trajectories and then use them for prediction. However, these learned MPs may not be indicative due to inadequate observation, which naturally affects the reliability of motion prediction. In this paper, we present an adaptive human motion analysis and prediction method. It adaptively predicts motion based on the classified MPs in terms of their credibility, which refers to how indicative the learned MPs are for the specific environment. The main contributions of the proposed method are as follows: First, it provides a comprehensive description of MPs including not only the learned MPs but also their evaluated credibility. Second, it predicts long-term future motion with reasonable accuracy. A number of experiments have been conducted in simulated scenes and real-world scenes and the prediction results have been quantitatively evaluated. The results show that the proposed method is effective and superior in its performance when compared with a recursively applied Auto-Regressive (AR) model, which is called the Recursive Short-term Predictor (RSP) for long-term prediction. The proposed method has 17.73% of improvement over the RSP in prediction accuracy in the experiment with the best performance. On average, the proposed method has 5% improvement over the RSP in prediction accuracy over 10 experiments. © 2011 Elsevier Ltd. All rights reserved.
 
dc.description.naturepostprint
 
dc.identifier.citationPattern Recognition, 2011, v. 44 n. 12, p. 2902-2914 [How to Cite?]
DOI: http://dx.doi.org/10.1016/j.patcog.2011.04.022
 
dc.identifier.citeulike9280508
 
dc.identifier.doihttp://dx.doi.org/10.1016/j.patcog.2011.04.022
 
dc.identifier.eissn1873-5142
 
dc.identifier.epage2914
 
dc.identifier.hkuros201090
 
dc.identifier.isiWOS:000292947000008
Funding AgencyGrant Number
Research Grant Council of the Hong Kong Special Administrative Region, ChinaHKU719406E
HKU719608E
University of Hong Kong
Funding Information:

The work on MP prediction was supported in part by a grant from the Research Grant Council of the Hong Kong Special Administrative Region, China, under Project HKU719406E and in part by the Postgraduate Studentship of The University of Hong Kong. The work on trajectory extraction was supported in part by a grant from the Research Grant Council of the Hong Kong Special Administrative Region, China, under Project HKU719608E and in part by the Postgraduate Studentship of the University of Hong Kong.

 
dc.identifier.issn0031-3203
2012 Impact Factor: 2.632
2012 SCImago Journal Rankings: 2.365
 
dc.identifier.issue12
 
dc.identifier.scopuseid_2-s2.0-79959362533
 
dc.identifier.spage2902
 
dc.identifier.urihttp://hdl.handle.net/10722/155622
 
dc.identifier.volume44
 
dc.languageeng
 
dc.publisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/pr
 
dc.publisher.placeNetherlands
 
dc.relation.ispartofPattern Recognition
 
dc.relation.referencesReferences in Scopus
 
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Pattern Recognition. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Pattern Recognition, 2011, v. 44 n. 12, p. 2902-2914. DOI: 10.1016/j.patcog.2011.04.022
 
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License
 
dc.subjectAuto regressive models
 
dc.subjectHuman motion analysis
 
dc.subjectHuman motions
 
dc.subjectLong-term future
 
dc.subjectLong-term prediction
 
dc.titleAdaptive human motion analysis and prediction
 
dc.typeArticle
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>Chen, Z</contributor.author>
<contributor.author>Wang, L</contributor.author>
<contributor.author>Yung, NHC</contributor.author>
<date.accessioned>2012-08-08T08:34:25Z</date.accessioned>
<date.available>2012-08-08T08:34:25Z</date.available>
<date.issued>2011</date.issued>
<identifier.citation>Pattern Recognition, 2011, v. 44 n. 12, p. 2902-2914</identifier.citation>
<identifier.issn>0031-3203</identifier.issn>
<identifier.uri>http://hdl.handle.net/10722/155622</identifier.uri>
<description.abstract>Human motion analysis and prediction is an active research area where predicting human motion is often performed for a single time step based on historical motion. In recent years, longer term human motion prediction has been attempted over a number of future time steps. Most current methods learn motion patterns (MPs) from observed trajectories and then use them for prediction. However, these learned MPs may not be indicative due to inadequate observation, which naturally affects the reliability of motion prediction. In this paper, we present an adaptive human motion analysis and prediction method. It adaptively predicts motion based on the classified MPs in terms of their credibility, which refers to how indicative the learned MPs are for the specific environment. The main contributions of the proposed method are as follows: First, it provides a comprehensive description of MPs including not only the learned MPs but also their evaluated credibility. Second, it predicts long-term future motion with reasonable accuracy. A number of experiments have been conducted in simulated scenes and real-world scenes and the prediction results have been quantitatively evaluated. The results show that the proposed method is effective and superior in its performance when compared with a recursively applied Auto-Regressive (AR) model, which is called the Recursive Short-term Predictor (RSP) for long-term prediction. The proposed method has 17.73% of improvement over the RSP in prediction accuracy in the experiment with the best performance. On average, the proposed method has 5% improvement over the RSP in prediction accuracy over 10 experiments. &#169; 2011 Elsevier Ltd. All rights reserved.</description.abstract>
<language>eng</language>
<publisher>Elsevier BV. The Journal&apos;s web site is located at http://www.elsevier.com/locate/pr</publisher>
<relation.ispartof>Pattern Recognition</relation.ispartof>
<rights>NOTICE: this is the author&#8217;s version of a work that was accepted for publication in Pattern Recognition. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Pattern Recognition, 2011, v. 44 n. 12, p. 2902-2914. DOI: 10.1016/j.patcog.2011.04.022</rights>
<rights>Creative Commons: Attribution 3.0 Hong Kong License</rights>
<subject>Auto regressive models</subject>
<subject>Human motion analysis</subject>
<subject>Human motions</subject>
<subject>Long-term future</subject>
<subject>Long-term prediction</subject>
<title>Adaptive human motion analysis and prediction</title>
<type>Article</type>
<description.nature>postprint</description.nature>
<identifier.doi>10.1016/j.patcog.2011.04.022</identifier.doi>
<identifier.scopus>eid_2-s2.0-79959362533</identifier.scopus>
<identifier.hkuros>201090</identifier.hkuros>
<relation.references>http://www.scopus.com/mlt/select.url?eid=2-s2.0-79959362533&amp;selection=ref&amp;src=s&amp;origin=recordpage</relation.references>
<identifier.volume>44</identifier.volume>
<identifier.issue>12</identifier.issue>
<identifier.spage>2902</identifier.spage>
<identifier.epage>2914</identifier.epage>
<identifier.eissn>1873-5142</identifier.eissn>
<identifier.isi>WOS:000292947000008</identifier.isi>
<publisher.place>Netherlands</publisher.place>
<identifier.citeulike>9280508</identifier.citeulike>
<bitstream.url>http://hub.hku.hk/bitstream/10722/155622/1/Content.pdf</bitstream.url>
</item>
Author Affiliations
  1. The University of Hong Kong