Unsupervised dance figure analysis from video for dancing avatar animation

Ferda Ofli, E. Erzin, Y. Yemez, A. M. Tekalp, Ç E. Erdem, A. T. Erdem, T. Abaci, M. K. Özkan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper presents a framework for unsupervised video analysis in the context of dance performances, where gestures and 3D movements of a dancer are characterized by repetition of a set of unknown dance figures. The system is trained in an unsupervised manner using Hidden Markov Models (HMMs) to automatically segment multi-view video recordings of a dancer into recurring elementary temporal body motion patterns to identify the dance figures. That is, a parallel HMM structure is employed to automatically determine the number and the temporal boundaries of different dance figures in a given dance video. The success of the analysis framework has been evaluated by visualizing these dance figures on a dancing avatar animated by the computed 3D analysis parameters. Experimental results demonstrate that the proposed framework enables synthetic agents and/or robots to learn dance figures from video automatically.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages1484-1487
Number of pages4
DOIs
Publication statusPublished - 1 Dec 2008
Externally publishedYes
Event2008 IEEE International Conference on Image Processing, ICIP 2008 - San Diego, CA, United States
Duration: 12 Oct 200815 Oct 2008

Other

Other2008 IEEE International Conference on Image Processing, ICIP 2008
CountryUnited States
CitySan Diego, CA
Period12/10/0815/10/08

Fingerprint

Hidden Markov models
Animation
Video recording
Model structures
Robots

Keywords

  • Dance figure identification
  • Dancing avatar animation
  • Unsupervised human body motion analysis

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Ofli, F., Erzin, E., Yemez, Y., Tekalp, A. M., Erdem, Ç. E., Erdem, A. T., ... Özkan, M. K. (2008). Unsupervised dance figure analysis from video for dancing avatar animation. In Proceedings - International Conference on Image Processing, ICIP (pp. 1484-1487). [4712047] https://doi.org/10.1109/ICIP.2008.4712047

Unsupervised dance figure analysis from video for dancing avatar animation. / Ofli, Ferda; Erzin, E.; Yemez, Y.; Tekalp, A. M.; Erdem, Ç E.; Erdem, A. T.; Abaci, T.; Özkan, M. K.

Proceedings - International Conference on Image Processing, ICIP. 2008. p. 1484-1487 4712047.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ofli, F, Erzin, E, Yemez, Y, Tekalp, AM, Erdem, ÇE, Erdem, AT, Abaci, T & Özkan, MK 2008, Unsupervised dance figure analysis from video for dancing avatar animation. in Proceedings - International Conference on Image Processing, ICIP., 4712047, pp. 1484-1487, 2008 IEEE International Conference on Image Processing, ICIP 2008, San Diego, CA, United States, 12/10/08. https://doi.org/10.1109/ICIP.2008.4712047
Ofli F, Erzin E, Yemez Y, Tekalp AM, Erdem ÇE, Erdem AT et al. Unsupervised dance figure analysis from video for dancing avatar animation. In Proceedings - International Conference on Image Processing, ICIP. 2008. p. 1484-1487. 4712047 https://doi.org/10.1109/ICIP.2008.4712047
Ofli, Ferda ; Erzin, E. ; Yemez, Y. ; Tekalp, A. M. ; Erdem, Ç E. ; Erdem, A. T. ; Abaci, T. ; Özkan, M. K. / Unsupervised dance figure analysis from video for dancing avatar animation. Proceedings - International Conference on Image Processing, ICIP. 2008. pp. 1484-1487
@inproceedings{77d1944cda674e6c9bd672aa55a03c5a,
title = "Unsupervised dance figure analysis from video for dancing avatar animation",
abstract = "This paper presents a framework for unsupervised video analysis in the context of dance performances, where gestures and 3D movements of a dancer are characterized by repetition of a set of unknown dance figures. The system is trained in an unsupervised manner using Hidden Markov Models (HMMs) to automatically segment multi-view video recordings of a dancer into recurring elementary temporal body motion patterns to identify the dance figures. That is, a parallel HMM structure is employed to automatically determine the number and the temporal boundaries of different dance figures in a given dance video. The success of the analysis framework has been evaluated by visualizing these dance figures on a dancing avatar animated by the computed 3D analysis parameters. Experimental results demonstrate that the proposed framework enables synthetic agents and/or robots to learn dance figures from video automatically.",
keywords = "Dance figure identification, Dancing avatar animation, Unsupervised human body motion analysis",
author = "Ferda Ofli and E. Erzin and Y. Yemez and Tekalp, {A. M.} and Erdem, {{\cC} E.} and Erdem, {A. T.} and T. Abaci and {\"O}zkan, {M. K.}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICIP.2008.4712047",
language = "English",
isbn = "1424417643",
pages = "1484--1487",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Unsupervised dance figure analysis from video for dancing avatar animation

AU - Ofli, Ferda

AU - Erzin, E.

AU - Yemez, Y.

AU - Tekalp, A. M.

AU - Erdem, Ç E.

AU - Erdem, A. T.

AU - Abaci, T.

AU - Özkan, M. K.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - This paper presents a framework for unsupervised video analysis in the context of dance performances, where gestures and 3D movements of a dancer are characterized by repetition of a set of unknown dance figures. The system is trained in an unsupervised manner using Hidden Markov Models (HMMs) to automatically segment multi-view video recordings of a dancer into recurring elementary temporal body motion patterns to identify the dance figures. That is, a parallel HMM structure is employed to automatically determine the number and the temporal boundaries of different dance figures in a given dance video. The success of the analysis framework has been evaluated by visualizing these dance figures on a dancing avatar animated by the computed 3D analysis parameters. Experimental results demonstrate that the proposed framework enables synthetic agents and/or robots to learn dance figures from video automatically.

AB - This paper presents a framework for unsupervised video analysis in the context of dance performances, where gestures and 3D movements of a dancer are characterized by repetition of a set of unknown dance figures. The system is trained in an unsupervised manner using Hidden Markov Models (HMMs) to automatically segment multi-view video recordings of a dancer into recurring elementary temporal body motion patterns to identify the dance figures. That is, a parallel HMM structure is employed to automatically determine the number and the temporal boundaries of different dance figures in a given dance video. The success of the analysis framework has been evaluated by visualizing these dance figures on a dancing avatar animated by the computed 3D analysis parameters. Experimental results demonstrate that the proposed framework enables synthetic agents and/or robots to learn dance figures from video automatically.

KW - Dance figure identification

KW - Dancing avatar animation

KW - Unsupervised human body motion analysis

UR - http://www.scopus.com/inward/record.url?scp=69949132134&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949132134&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2008.4712047

DO - 10.1109/ICIP.2008.4712047

M3 - Conference contribution

AN - SCOPUS:69949132134

SN - 1424417643

SN - 9781424417643

SP - 1484

EP - 1487

BT - Proceedings - International Conference on Image Processing, ICIP

ER -