Searching for Variable-Speed Motions in Long Sequences of
Motion Capture Data

J 2019

Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA

Basic information

Original name

Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data

Authors

SEDMIDUBSKÝ, Jan (203 Czech Republic, guarantor, belonging to the institution), Petr ELIÁŠ (203 Czech Republic, belonging to the institution) and Pavel ZEZULA (203 Czech Republic, belonging to the institution)

Edition

Information Systems, 2019, 0306-4379

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

10200 1.2 Computer and information sciences

Country of publisher

Netherlands

Confidentiality degree

není předmětem státního či obchodního tajemství

References:

URL

Impact factor

Impact factor: 2.466

RIV identification code

RIV/00216224:14330/19:00107162

Organization unit

Faculty of Informatics

DOI

http://dx.doi.org/10.1016/j.is.2018.04.002

UT WoS

000454964800012

Keywords in English

Content-based retrieval;Motion capture data;Subsequence matching;Speed-invariant retrieval;Similarity measure;Hierarchical segmentation;Indexing

Abstract

V originále

Motion capture data digitally represent human movements by sequences of body configurations in time. Subsequence searching in long sequences of such spatio-temporal data is difficult as query-relevant motions can vary in execution speeds and styles and can occur anywhere in a very long data sequence. To deal with these problems, we employ a fast and effective similarity measure that is elastic. The property of elasticity enables matching of two overlapping but slightly misaligned subsequences with a high confidence. Based on the elasticity, the long data sequence is partitioned into overlapping segments that are organized in multiple levels. The number of levels and sizes of overlaps are optimized to generate a modest number of segments while being able to trace an arbitrary query. In a retrieval phase, a query is always represented as a single segment and fast matched against segments within a relevant level without any costly post-processing. Moreover, visiting adjacent levels makes possible subsequence searching of time-warped (i.e., faster or slower executed) queries. To efficiently search on a large scale, segment features can be binarized and segmentation levels independently indexed. We experimentally demonstrate effectiveness and efficiency of the proposed approach for subsequence searching on a real-life dataset.

Links

GA16-18889S, research and development project

Name: Analytika pro velká nestrukturovaná data (Acronym: Big Data Analytics for Unstructured Data)

Investor: Czech Science Foundation

Citovat

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA. Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data. Information Systems. 2019, vol. 80, February, p. 148-158. ISSN 0306-4379. Available from: https://dx.doi.org/10.1016/j.is.2018.04.002.

@article{1413841,
   author = {Sedmidubský, Jan and Eliáš, Petr and Zezula, Pavel},
   article_number = {February},
   doi = {http://dx.doi.org/10.1016/j.is.2018.04.002},
   keywords = {Content-based retrieval;Motion capture data;Subsequence matching;Speed-invariant retrieval;Similarity measure;Hierarchical segmentation;Indexing},
   language = {eng},
   issn = {0306-4379},
   journal = {Information Systems},
   title = {Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data},
   url = {http://dx.doi.org/10.1016/j.is.2018.04.002},
   volume = {80},
   year = {2019}
}

TY  - JOUR
ID  - 1413841
AU  - Sedmidubský, Jan - Eliáš, Petr - Zezula, Pavel
PY  - 2019
TI  - Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data
JF  - Information Systems
VL  - 80
IS  - February
SP  - 148-158
EP  - 148-158
SN  - 03064379
KW  - Content-based retrieval;Motion capture data;Subsequence matching;Speed-invariant retrieval;Similarity measure;Hierarchical segmentation;Indexing
UR  - http://dx.doi.org/10.1016/j.is.2018.04.002
L2  - http://dx.doi.org/10.1016/j.is.2018.04.002
N2  - Motion capture data digitally represent human movements by sequences of body configurations in time. Subsequence searching in long sequences of such spatio-temporal data is difficult as query-relevant motions can vary in execution speeds and styles and can occur anywhere in a very long data sequence. To deal with these problems, we employ a fast and effective similarity measure that is elastic. The property of elasticity enables matching of two overlapping but slightly misaligned subsequences with a high confidence. Based on the elasticity, the long data sequence is partitioned into overlapping segments that are organized in multiple levels. The number of levels and sizes of overlaps are optimized to generate a modest number of segments while being able to trace an arbitrary query. In a retrieval phase, a query is always represented as a single segment and fast matched against segments within a relevant level without any costly post-processing. Moreover, visiting adjacent levels makes possible subsequence searching of time-warped (i.e., faster or slower executed) queries. To efficiently search on a large scale, segment features can be binarized and segmentation levels independently indexed. We experimentally demonstrate effectiveness and efficiency of the proposed approach for subsequence searching on a real-life dataset.
ER  -

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA. Searching for Variable-Speed Motions in Long Sequences of Motion Capture Data. \textit{Information Systems}. 2019, vol.~80, February, p.~148-158. ISSN~0306-4379. Available from: https://dx.doi.org/10.1016/j.is.2018.04.002.

Detailed Information on Publication Record