Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases

Christian Beecks; M. Hassani; F. Obeloer; Thomas Seidl

doi:10.1109/ICDMW.2015.194

Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases

Christian Beecks, M. Hassani, F. Obeloer, Thomas Seidl

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

3 Citations (Scopus)

Abstract

One of the most fundamental challenges when mining gestural patterns in 3D motion capture databases is the definition of spatiotemporal similarity between two gestural patterns. While time-elastic similarity models such as the Gesture Matching Distance on gesture signatures are able to leverage the spatial and temporal characteristics of gestural patterns, the applicability of such distance-based models in order to analyze large 3D motion capture databases is limited due to their high computational complexity. To this end, we propose a lower bound approximation of the Gesture Matching Distance that preserves the spatiotemporal characteristics and can be utilized in an optimal multi-step k-nearest-neighbor search architecture in order to analyze and mine spatiotemporal databases efficiently. We empirically investigate the performance in terms of accuracy and efficiency based on 3D motion capture databases and show that our lower bound approximation is able to achieve an increase in efficiency of more than one order of magnitude with a negligible loss in accuracy. Our proposal is fundamental for efficient distance-based gestural pattern mining.

Original language	English
Title of host publication	15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings
Editors	P. Cui, J. Dy, C. Aggarwal, Z.-H. Zhou, A. Tuzhilin, H. Xiong, X. Wu
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Pages	1425-1432
Number of pages	8
ISBN (Electronic)	978-1-4673-8493-3
ISBN (Print)	9781467384926
DOIs	https://doi.org/10.1109/ICDMW.2015.194
Publication status	Published - 29 Jan 2016
Externally published	Yes
Event	15th IEEE International Conference on Data Mining (ICDM 2015) - Atlantic City, United States Duration: 14 Nov 2015 → 17 Nov 2015 Conference number: 15 https://icdm2015.stonybrook.edu/

Conference

Conference	15th IEEE International Conference on Data Mining (ICDM 2015)
Abbreviated title	ICDM 2015
Country/Territory	United States
City	Atlantic City
Period	14/11/15 → 17/11/15
Internet address	https://icdm2015.stonybrook.edu/

Keywords

3D motion capture data
Gesture matching distance
gestural pattern mining
gesture signature
spatiotemporal similarity search

Access to Document

10.1109/ICDMW.2015.194

http://www.mendeley.com/research/efficient-distancebased-gestural-pattern-mining-spatiotemporal-3d-motion-capture-databases

Cite this

Beecks, C., Hassani, M., Obeloer, F., & Seidl, T. (2016). Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases. In P. Cui, J. Dy, C. Aggarwal, Z.-H. Zhou, A. Tuzhilin, H. Xiong, & X. Wu (Eds.), 15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings (pp. 1425-1432). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICDMW.2015.194

Beecks, Christian ; Hassani, M. ; Obeloer, F. et al. / Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases. 15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings. editor / P. Cui ; J. Dy ; C. Aggarwal ; Z.-H. Zhou ; A. Tuzhilin ; H. Xiong ; X. Wu. Piscataway : Institute of Electrical and Electronics Engineers, 2016. pp. 1425-1432

@inproceedings{228e78e2b01344c6a9215dc62ea32a53,

title = "Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases",

abstract = "One of the most fundamental challenges when mining gestural patterns in 3D motion capture databases is the definition of spatiotemporal similarity between two gestural patterns. While time-elastic similarity models such as the Gesture Matching Distance on gesture signatures are able to leverage the spatial and temporal characteristics of gestural patterns, the applicability of such distance-based models in order to analyze large 3D motion capture databases is limited due to their high computational complexity. To this end, we propose a lower bound approximation of the Gesture Matching Distance that preserves the spatiotemporal characteristics and can be utilized in an optimal multi-step k-nearest-neighbor search architecture in order to analyze and mine spatiotemporal databases efficiently. We empirically investigate the performance in terms of accuracy and efficiency based on 3D motion capture databases and show that our lower bound approximation is able to achieve an increase in efficiency of more than one order of magnitude with a negligible loss in accuracy. Our proposal is fundamental for efficient distance-based gestural pattern mining.",

keywords = "3D motion capture data, Gesture matching distance, gestural pattern mining, gesture signature, spatiotemporal similarity search",

author = "Christian Beecks and M. Hassani and F. Obeloer and Thomas Seidl",

year = "2016",

month = jan,

day = "29",

doi = "10.1109/ICDMW.2015.194",

language = "English",

isbn = "9781467384926",

pages = "1425--1432",

editor = "P. Cui and J. Dy and C. Aggarwal and Z.-H. Zhou and A. Tuzhilin and H. Xiong and X. Wu",

booktitle = "15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "15th IEEE International Conference on Data Mining (ICDM 2015), ICDM 2015 ; Conference date: 14-11-2015 Through 17-11-2015",

url = "https://icdm2015.stonybrook.edu/",

}

Beecks, C, Hassani, M, Obeloer, F & Seidl, T 2016, Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases. in P Cui, J Dy, C Aggarwal, Z-H Zhou, A Tuzhilin, H Xiong & X Wu (eds), 15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings. Institute of Electrical and Electronics Engineers, Piscataway, pp. 1425-1432, 15th IEEE International Conference on Data Mining (ICDM 2015), Atlantic City, Georgia, United States, 14/11/15. https://doi.org/10.1109/ICDMW.2015.194

Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases. / Beecks, Christian; Hassani, M.; Obeloer, F. et al.
15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings. ed. / P. Cui; J. Dy; C. Aggarwal; Z.-H. Zhou; A. Tuzhilin; H. Xiong; X. Wu. Piscataway: Institute of Electrical and Electronics Engineers, 2016. p. 1425-1432.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases

AU - Beecks, Christian

AU - Hassani, M.

AU - Obeloer, F.

AU - Seidl, Thomas

N1 - Conference code: 15

PY - 2016/1/29

Y1 - 2016/1/29

N2 - One of the most fundamental challenges when mining gestural patterns in 3D motion capture databases is the definition of spatiotemporal similarity between two gestural patterns. While time-elastic similarity models such as the Gesture Matching Distance on gesture signatures are able to leverage the spatial and temporal characteristics of gestural patterns, the applicability of such distance-based models in order to analyze large 3D motion capture databases is limited due to their high computational complexity. To this end, we propose a lower bound approximation of the Gesture Matching Distance that preserves the spatiotemporal characteristics and can be utilized in an optimal multi-step k-nearest-neighbor search architecture in order to analyze and mine spatiotemporal databases efficiently. We empirically investigate the performance in terms of accuracy and efficiency based on 3D motion capture databases and show that our lower bound approximation is able to achieve an increase in efficiency of more than one order of magnitude with a negligible loss in accuracy. Our proposal is fundamental for efficient distance-based gestural pattern mining.

AB - One of the most fundamental challenges when mining gestural patterns in 3D motion capture databases is the definition of spatiotemporal similarity between two gestural patterns. While time-elastic similarity models such as the Gesture Matching Distance on gesture signatures are able to leverage the spatial and temporal characteristics of gestural patterns, the applicability of such distance-based models in order to analyze large 3D motion capture databases is limited due to their high computational complexity. To this end, we propose a lower bound approximation of the Gesture Matching Distance that preserves the spatiotemporal characteristics and can be utilized in an optimal multi-step k-nearest-neighbor search architecture in order to analyze and mine spatiotemporal databases efficiently. We empirically investigate the performance in terms of accuracy and efficiency based on 3D motion capture databases and show that our lower bound approximation is able to achieve an increase in efficiency of more than one order of magnitude with a negligible loss in accuracy. Our proposal is fundamental for efficient distance-based gestural pattern mining.

KW - 3D motion capture data

KW - Gesture matching distance

KW - gestural pattern mining

KW - gesture signature

KW - spatiotemporal similarity search

U2 - 10.1109/ICDMW.2015.194

DO - 10.1109/ICDMW.2015.194

M3 - Conference contribution

SN - 9781467384926

SP - 1425

EP - 1432

BT - 15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings

A2 - Cui, P.

A2 - Dy, J.

A2 - Aggarwal, C.

A2 - Zhou, Z.-H.

A2 - Tuzhilin, A.

A2 - Xiong, H.

A2 - Wu, X.

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 15th IEEE International Conference on Data Mining (ICDM 2015)

Y2 - 14 November 2015 through 17 November 2015

ER -

Beecks C, Hassani M, Obeloer F, Seidl T. Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases. In Cui P, Dy J, Aggarwal C, Zhou ZH, Tuzhilin A, Xiong H, Wu X, editors, 15th IEEE International Conference on Data Mining workshop 14-17 November 2015, Atlantic City, New Jersey : Proceedings. Piscataway: Institute of Electrical and Electronics Engineers. 2016. p. 1425-1432 doi: 10.1109/ICDMW.2015.194

Efficient distance-based gestural pattern mining in spatiotemporal 3D motion capture databases

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this