Laplacian one class extreme learning machines for human action recognition

Vasileios Mygdalis; Alexandros Iosifidis; Anastasios Tefas; Ioannis Pitas

doi:10.1109/MMSP.2016.7813387

Laplacian one class extreme learning machines for human action recognition

Vasileios Mygdalis, Alexandros Iosifidis, Anastasios Tefas, Ioannis Pitas

Department of Electrical & Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

2 Citations (Scopus)

245 Downloads (Pure)

Abstract

A novel OCC method for human action recognition namely the Laplacian One Class Extreme Learning Machines is presented. The proposed method exploits local geometric data information within the OC-ELM optimization process. It is shown that emphasizing on preserving the local geometry of the data leads to a regularized solution, which models the target class more efficiently than the standard OC-ELM algorithm. The proposed method is extended to operate in feature spaces determined by the network hidden layer outputs, as well as in ELM spaces of arbitrary dimensions. Its superior performance against other OCC options is consistent among five publicly available human action recognition datasets.

Original language	English
Title of host publication	2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016)
Subtitle of host publication	Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	298-302
Number of pages	5
ISBN (Electronic)	9781509037247
ISBN (Print)	9781509037254
DOIs	https://doi.org/10.1109/MMSP.2016.7813387
Publication status	Published - Feb 2017
Event	18th International Workshop on Multimedia Signal Processing: MMSP2016 - Montreal, Canada Duration: 21 Sep 2016 → 23 Sep 2016

Publication series

Name	Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
ISSN (Print)	2163-3517
ISSN (Electronic)	2473-3628

Conference

Conference	18th International Workshop on Multimedia Signal Processing
Country	Canada
City	Montreal
Period	21/09/16 → 23/09/16

Access to Document

10.1109/MMSP.2016.7813387

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/document/7813387/. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 3.45 MB

Fingerprint Dive into the research topics of 'Laplacian one class extreme learning machines for human action recognition'. Together they form a unique fingerprint.

Cite this

Mygdalis, V., Iosifidis, A., Tefas, A., & Pitas, I. (2017). Laplacian one class extreme learning machines for human action recognition. In 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016): Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada (pp. 298-302). (Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/MMSP.2016.7813387

Mygdalis, Vasileios ; Iosifidis, Alexandros ; Tefas, Anastasios ; Pitas, Ioannis. / Laplacian one class extreme learning machines for human action recognition. 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016): Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada. Institute of Electrical and Electronics Engineers (IEEE), 2017. pp. 298-302 (Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)).

@inproceedings{b475a7189f14441f871efa2fcf08ff0a,

title = "Laplacian one class extreme learning machines for human action recognition",

abstract = "A novel OCC method for human action recognition namely the Laplacian One Class Extreme Learning Machines is presented. The proposed method exploits local geometric data information within the OC-ELM optimization process. It is shown that emphasizing on preserving the local geometry of the data leads to a regularized solution, which models the target class more efficiently than the standard OC-ELM algorithm. The proposed method is extended to operate in feature spaces determined by the network hidden layer outputs, as well as in ELM spaces of arbitrary dimensions. Its superior performance against other OCC options is consistent among five publicly available human action recognition datasets.",

author = "Vasileios Mygdalis and Alexandros Iosifidis and Anastasios Tefas and Ioannis Pitas",

year = "2017",

month = feb,

doi = "10.1109/MMSP.2016.7813387",

language = "English",

isbn = "9781509037254",

series = "Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "298--302",

booktitle = "2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016)",

address = "United States",

note = " 18th International Workshop on Multimedia Signal Processing : MMSP2016 ; Conference date: 21-09-2016 Through 23-09-2016",

}

Mygdalis, V, Iosifidis, A, Tefas, A & Pitas, I 2017, Laplacian one class extreme learning machines for human action recognition. in 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016): Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada. Proceedings of the International Workshop on Multimedia Signal Processing (MMSP), Institute of Electrical and Electronics Engineers (IEEE), pp. 298-302, 18th International Workshop on Multimedia Signal Processing, Montreal, Canada, 21/09/16. https://doi.org/10.1109/MMSP.2016.7813387

Laplacian one class extreme learning machines for human action recognition. / Mygdalis, Vasileios; Iosifidis, Alexandros; Tefas, Anastasios; Pitas, Ioannis.

2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016): Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada. Institute of Electrical and Electronics Engineers (IEEE), 2017. p. 298-302 (Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)).

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Laplacian one class extreme learning machines for human action recognition

AU - Mygdalis, Vasileios

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Pitas, Ioannis

PY - 2017/2

Y1 - 2017/2

N2 - A novel OCC method for human action recognition namely the Laplacian One Class Extreme Learning Machines is presented. The proposed method exploits local geometric data information within the OC-ELM optimization process. It is shown that emphasizing on preserving the local geometry of the data leads to a regularized solution, which models the target class more efficiently than the standard OC-ELM algorithm. The proposed method is extended to operate in feature spaces determined by the network hidden layer outputs, as well as in ELM spaces of arbitrary dimensions. Its superior performance against other OCC options is consistent among five publicly available human action recognition datasets.

AB - A novel OCC method for human action recognition namely the Laplacian One Class Extreme Learning Machines is presented. The proposed method exploits local geometric data information within the OC-ELM optimization process. It is shown that emphasizing on preserving the local geometry of the data leads to a regularized solution, which models the target class more efficiently than the standard OC-ELM algorithm. The proposed method is extended to operate in feature spaces determined by the network hidden layer outputs, as well as in ELM spaces of arbitrary dimensions. Its superior performance against other OCC options is consistent among five publicly available human action recognition datasets.

U2 - 10.1109/MMSP.2016.7813387

DO - 10.1109/MMSP.2016.7813387

M3 - Conference Contribution (Conference Proceeding)

SN - 9781509037254

T3 - Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)

SP - 298

EP - 302

BT - 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 18th International Workshop on Multimedia Signal Processing

Y2 - 21 September 2016 through 23 September 2016

ER -

Mygdalis V, Iosifidis A, Tefas A, Pitas I. Laplacian one class extreme learning machines for human action recognition. In 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP 2016): Proceedings of a meeting held 21-23 September 2016, Montreal, QC, Canada. Institute of Electrical and Electronics Engineers (IEEE). 2017. p. 298-302. (Proceedings of the International Workshop on Multimedia Signal Processing (MMSP)). https://doi.org/10.1109/MMSP.2016.7813387