A System for analyzing and indexing human-motion databases

Guodong Liu, Jingdan Zhang, Wei Wang, Leonard McMillan

Research output: Contribution to journalConference article

65 Citations (Scopus)

Abstract

We demonstrate a data-driven approach for representing, compressing, and indexing human-motion databases. Our modeling approach is based on piecewise-linear components that are determined via a divisive clustering method. Selection of the appropriate linear model is determined automatically via a classifier using a subspace of the most significant, or principle features (markers). We show that, after offline training, our model can accurately estimate and classify human motions. We can also construct indexing structures for motion sequences according to their transition trajectories through these linear components. Our method not only provides indices for whole and/or partial motion sequences, but also serves as a compressed representation for the entire motion database. Our method also tends to be immune to tremporal variations, and thus avoids the expense of time-warping.

Original languageEnglish (US)
Pages (from-to)924-926
Number of pages3
JournalProceedings of the ACM SIGMOD International Conference on Management of Data
StatePublished - Dec 1 2005
EventSIGMOD 2005: ACM SIGMOD International Conference on Management of Data - Baltimore, MD, United States
Duration: Jun 14 2005Jun 16 2005

Fingerprint

Classifiers
Trajectories

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems

Cite this

@article{1df240c7e81245f9a3b09256fbf4f274,
title = "A System for analyzing and indexing human-motion databases",
abstract = "We demonstrate a data-driven approach for representing, compressing, and indexing human-motion databases. Our modeling approach is based on piecewise-linear components that are determined via a divisive clustering method. Selection of the appropriate linear model is determined automatically via a classifier using a subspace of the most significant, or principle features (markers). We show that, after offline training, our model can accurately estimate and classify human motions. We can also construct indexing structures for motion sequences according to their transition trajectories through these linear components. Our method not only provides indices for whole and/or partial motion sequences, but also serves as a compressed representation for the entire motion database. Our method also tends to be immune to tremporal variations, and thus avoids the expense of time-warping.",
author = "Guodong Liu and Jingdan Zhang and Wei Wang and Leonard McMillan",
year = "2005",
month = "12",
day = "1",
language = "English (US)",
pages = "924--926",
journal = "Proceedings of the ACM SIGMOD International Conference on Management of Data",
issn = "0730-8078",
publisher = "Association for Computing Machinery (ACM)",

}

A System for analyzing and indexing human-motion databases. / Liu, Guodong; Zhang, Jingdan; Wang, Wei; McMillan, Leonard.

In: Proceedings of the ACM SIGMOD International Conference on Management of Data, 01.12.2005, p. 924-926.

Research output: Contribution to journalConference article

TY - JOUR

T1 - A System for analyzing and indexing human-motion databases

AU - Liu, Guodong

AU - Zhang, Jingdan

AU - Wang, Wei

AU - McMillan, Leonard

PY - 2005/12/1

Y1 - 2005/12/1

N2 - We demonstrate a data-driven approach for representing, compressing, and indexing human-motion databases. Our modeling approach is based on piecewise-linear components that are determined via a divisive clustering method. Selection of the appropriate linear model is determined automatically via a classifier using a subspace of the most significant, or principle features (markers). We show that, after offline training, our model can accurately estimate and classify human motions. We can also construct indexing structures for motion sequences according to their transition trajectories through these linear components. Our method not only provides indices for whole and/or partial motion sequences, but also serves as a compressed representation for the entire motion database. Our method also tends to be immune to tremporal variations, and thus avoids the expense of time-warping.

AB - We demonstrate a data-driven approach for representing, compressing, and indexing human-motion databases. Our modeling approach is based on piecewise-linear components that are determined via a divisive clustering method. Selection of the appropriate linear model is determined automatically via a classifier using a subspace of the most significant, or principle features (markers). We show that, after offline training, our model can accurately estimate and classify human motions. We can also construct indexing structures for motion sequences according to their transition trajectories through these linear components. Our method not only provides indices for whole and/or partial motion sequences, but also serves as a compressed representation for the entire motion database. Our method also tends to be immune to tremporal variations, and thus avoids the expense of time-warping.

UR - http://www.scopus.com/inward/record.url?scp=29844443064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=29844443064&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:29844443064

SP - 924

EP - 926

JO - Proceedings of the ACM SIGMOD International Conference on Management of Data

JF - Proceedings of the ACM SIGMOD International Conference on Management of Data

SN - 0730-8078

ER -