Benchmarking Search and Annotation in Continuous Human Skeleton
Sequences

D 2019

Benchmarking Search and Annotation in Continuous Human Skeleton Sequences

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA

Basic information

Original name

Benchmarking Search and Annotation in Continuous Human Skeleton Sequences

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

New York, NY, USA, International Conference on Multimedia Retrieval (ICMR), p. 38-42, 5 pp. 2019

Publisher

ACM

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10200 1.2 Computer and information sciences

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

RIV identification code

RIV/00216224:14330/19:00107371

Organization unit

Faculty of Informatics

ISBN

978-1-4503-6765-3

DOI

http://dx.doi.org/10.1145/3323873.3325013

UT WoS

000482188900008

Keywords in English

motion capture dataset;continuous 3D skeleton sequence;stream-based processing;benchmark;subsequence search;action detection;mining

Abstract

V originále

Motion capture data are digital representations of human movements in form of 3D trajectories of multiple body joints. To understand the captured motions, similarity-based processing and deep learning have already proved to be effective, especially in classifying pre-segmented actions. However, in real-world scenarios motion data are typically captured as long continuous sequences, without explicit knowledge of semantic partitioning. To make such unsegmented data accessible and reusable as required by many applications, there is a strong requirement to analyze, search, annotate and mine them automatically. However, there is currently an absence of datasets and benchmarks to test and compare the capabilities of the developed techniques for continuous motion data processing. In this paper, we introduce a new large-scale LSMB19 dataset consisting of two 3D skeleton sequences of a total length of 54.5 hours. We also define a benchmark on two important multimedia retrieval operations: subsequence search and annotation. Additionally, we exemplify the usability of the benchmark by establishing baseline results for these operations.

Links

GA19-02033S, research and development project

Name: Vyhledávání, analytika a anotace datových toků lidských pohybů

Investor: Czech Science Foundation

Citovat

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA. Benchmarking Search and Annotation in Continuous Human Skeleton Sequences. Online. In International Conference on Multimedia Retrieval (ICMR). New York, NY, USA: ACM, 2019, p. 38-42. ISBN 978-1-4503-6765-3. Available from: https://dx.doi.org/10.1145/3323873.3325013.

@inproceedings{1525496,
   author = {Sedmidubský, Jan and Eliáš, Petr and Zezula, Pavel},
   address = {New York, NY, USA},
   booktitle = {International Conference on Multimedia Retrieval (ICMR)},
   doi = {http://dx.doi.org/10.1145/3323873.3325013},
   keywords = {motion capture dataset;continuous 3D skeleton sequence;stream-based processing;benchmark;subsequence search;action detection;mining},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {New York, NY, USA},
   isbn = {978-1-4503-6765-3},
   pages = {38-42},
   publisher = {ACM},
   title = {Benchmarking Search and Annotation in Continuous Human Skeleton Sequences},
   year = {2019}
}

TY  - JOUR
ID  - 1525496
AU  - Sedmidubský, Jan - Eliáš, Petr - Zezula, Pavel
PY  - 2019
TI  - Benchmarking Search and Annotation in Continuous Human Skeleton Sequences
PB  - ACM
CY  - New York, NY, USA
SN  - 9781450367653
KW  - motion capture dataset;continuous 3D skeleton sequence;stream-based processing;benchmark;subsequence search;action detection;mining
N2  - Motion capture data are digital representations of human movements in form of 3D trajectories of multiple body joints. To understand the captured motions, similarity-based processing and deep learning have already proved to be effective, especially in classifying pre-segmented actions. However, in real-world scenarios motion data are typically captured as long continuous sequences, without explicit knowledge of semantic partitioning. To make such unsegmented data accessible and reusable as required by many applications, there is a strong requirement to analyze, search, annotate and mine them automatically. However, there is currently an absence of datasets and benchmarks to test and compare the capabilities of the developed techniques for continuous motion data processing. In this paper, we introduce a new large-scale LSMB19 dataset consisting of two 3D skeleton sequences of a total length of 54.5 hours. We also define a benchmark on two important multimedia retrieval operations: subsequence search and annotation. Additionally, we exemplify the usability of the benchmark by establishing baseline results for these operations.
ER  -

SEDMIDUBSKÝ, Jan, Petr ELIÁŠ and Pavel ZEZULA. Benchmarking Search and Annotation in Continuous Human Skeleton Sequences. Online. In \textit{International Conference on Multimedia Retrieval (ICMR)}. New York, NY, USA: ACM, 2019, p.~38-42. ISBN~978-1-4503-6765-3. Available from: https://dx.doi.org/10.1145/3323873.3325013.

Detailed Information on Publication Record