Probabilistic Classification of Skeleton Sequences

D 2018

Probabilistic Classification of Skeleton Sequences

SEDMIDUBSKÝ, Jan and Pavel ZEZULA

Basic information

Original name

Probabilistic Classification of Skeleton Sequences

Authors

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

Edition

Switzerland, 29th International Conference on Database and Expert Systems Applications (DEXA 2018), p. 50-65, 16 pp. 2018

Publisher

Springer

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

storage medium (CD, DVD, flash disk)

Impact factor

Impact factor: 0.402 in 2005

RIV identification code

RIV/00216224:14330/18:00100948

Organization unit

Faculty of Informatics

ISBN

978-3-319-98811-5

ISSN

DOI

http://dx.doi.org/10.1007/978-3-319-98812-2_4

UT WoS

000460551600004

Keywords in English

motion capture data; nearest-neighbor search; action recognition; action classification; re-ranking; similarity measure

Abstract

V originále

Automatic classification of 3D skeleton sequences of human motions has applications in many domains, ranging from entertainment to medicine. The classification is a difficult problem as the motions belonging to the same class needn't be well segmented and can be performed by subjects of various body sizes in different styles and speeds. The state-of-the-art recognition approaches commonly solve this problem by training recurrent neural networks to learn the contextual dependency in both spatial and temporal domains. In this paper, we employ a distance-based similarity measure, based on deep convolutional features, to search for the k-nearest motions with respect to a query motion being classified. The retrieved neighbors are analyzed and re-ranked by additional measures that are automatically chosen for individual queries. The combination of deep features, dynamism in the similarity-measure selection, and a new kNN classifier brings the highest classification accuracy on a challenging dataset with 130 classes. Moreover, the proposed approach can promptly react to changing training data without any need for a retraining process.

Links

GBP103/12/G084, research and development project

Name: Centrum pro multi-modální interpretaci dat velkého rozsahu

Investor: Czech Science Foundation

Citovat

SEDMIDUBSKÝ, Jan and Pavel ZEZULA. Probabilistic Classification of Skeleton Sequences. In S. Hartmann et al. 29th International Conference on Database and Expert Systems Applications (DEXA 2018). Switzerland: Springer, 2018, p. 50-65. ISBN 978-3-319-98811-5. Available from: https://dx.doi.org/10.1007/978-3-319-98812-2_4.

@inproceedings{1417925,
   author = {Sedmidubský, Jan and Zezula, Pavel},
   address = {Switzerland},
   booktitle = {29th International Conference on Database and Expert Systems Applications (DEXA 2018)},
   doi = {http://dx.doi.org/10.1007/978-3-319-98812-2_4},
   editor = {S. Hartmann et al.},
   keywords = {motion capture data; nearest-neighbor search; action recognition; action classification; re-ranking; similarity measure},
   howpublished = {paměťový nosič},
   language = {eng},
   location = {Switzerland},
   isbn = {978-3-319-98811-5},
   pages = {50-65},
   publisher = {Springer},
   title = {Probabilistic Classification of Skeleton Sequences},
   year = {2018}
}

TY  - JOUR
ID  - 1417925
AU  - Sedmidubský, Jan - Zezula, Pavel
PY  - 2018
TI  - Probabilistic Classification of Skeleton Sequences
PB  - Springer
CY  - Switzerland
SN  - 9783319988115
KW  - motion capture data
KW  - nearest-neighbor search
KW  - action recognition
KW  - action classification
KW  - re-ranking
KW  - similarity measure
N2  - Automatic classification of 3D skeleton sequences of human motions has applications in many domains, ranging from entertainment to medicine. The classification is a difficult problem as the motions belonging to the same class needn't be well segmented and can be performed by subjects of various body sizes in different styles and speeds. The state-of-the-art recognition approaches commonly solve this problem by training recurrent neural networks to learn the contextual dependency in both spatial and temporal domains. In this paper, we employ a distance-based similarity measure, based on deep convolutional features, to search for the k-nearest motions with respect to a query motion being classified. The retrieved neighbors are analyzed and re-ranked by additional measures that are automatically chosen for individual queries. The combination of deep features, dynamism in the similarity-measure selection, and a new kNN classifier brings the highest classification accuracy on a challenging dataset with 130 classes. Moreover, the proposed approach can promptly react to changing training data without any need for a retraining process.
ER  -

SEDMIDUBSKÝ, Jan and Pavel ZEZULA. Probabilistic Classification of Skeleton Sequences. In S. Hartmann et al. \textit{29th International Conference on Database and Expert Systems Applications (DEXA 2018)}. Switzerland: Springer, 2018, p.~50-65. ISBN~978-3-319-98811-5. Available from: https://dx.doi.org/10.1007/978-3-319-98812-2\_{}4.

Detailed Information on Publication Record