Semi-supervised classification of human actions based on Neural Networks

Alexandros Iosifidis; Anastasios Tefas; Ioannis Pitas

doi:10.1109/ICPR.2014.239

Semi-supervised classification of human actions based on Neural Networks

Alexandros Iosifidis
, Anastasios Tefas
, Ioannis Pitas

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

21 Citations (Scopus)

423 Downloads (Pure)

Abstract

In this paper, we propose a novel algorithm for Single-hidden Layer Feedforward Neural networks training which is able to exploit information coming from both labeled and unlabeled data for semi-supervised action classification. We extend
the Extreme Learning Machine algorithm by incorporating appropriate regularization terms describing geometric properties and discrimination criteria of the training data representation in the ELM space to this end. The proposed algorithm is evaluated on human action recognition, where its performance is compared with that of other (semi-)supervised classification schemes. Experimental results on two publicly available action recognition databases denote its effectiveness.

Original language	English
Title of host publication	2014 22nd International Conference on Pattern Recognition (ICPR 2014)
Subtitle of host publication	Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1336-1341
Number of pages	6
ISBN (Print)	9781479952106
DOIs	https://doi.org/10.1109/ICPR.2014.239
Publication status	Published - Jan 2015
Event	IEEE International Conference on Pattern Recognition (ICPR) - Stockholm, Sweden Duration: 24 Aug 2014 → 28 Aug 2014

Publication series

Name	Proceedings of the International Conference on Pattern Recognition (ICPR)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
ISSN (Print)	1051-4651

Conference

Conference	IEEE International Conference on Pattern Recognition (ICPR)
Country/Territory	Sweden
City	Stockholm
Period	24/08/14 → 28/08/14

Access to Document

10.1109/ICPR.2014.239

Ionnas Pitas - Semi-supervised classification of human actions based on Neural Networks - 2014
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6976949. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 2.61 MB

Handle.net

Persistent link

Cite this

Iosifidis, A., Tefas, A., & Pitas, I. (2015). Semi-supervised classification of human actions based on Neural Networks. In 2014 22nd International Conference on Pattern Recognition (ICPR 2014): Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden (pp. 1336-1341). (Proceedings of the International Conference on Pattern Recognition (ICPR)). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ICPR.2014.239

Iosifidis, Alexandros ; Tefas, Anastasios ; Pitas, Ioannis. / Semi-supervised classification of human actions based on Neural Networks. 2014 22nd International Conference on Pattern Recognition (ICPR 2014): Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE), 2015. pp. 1336-1341 (Proceedings of the International Conference on Pattern Recognition (ICPR)).

@inproceedings{3b49fa2e0c4e407ab2c6e38e69510416,

title = "Semi-supervised classification of human actions based on Neural Networks",

abstract = "In this paper, we propose a novel algorithm for Single-hidden Layer Feedforward Neural networks training which is able to exploit information coming from both labeled and unlabeled data for semi-supervised action classification. We extend the Extreme Learning Machine algorithm by incorporating appropriate regularization terms describing geometric properties and discrimination criteria of the training data representation in the ELM space to this end. The proposed algorithm is evaluated on human action recognition, where its performance is compared with that of other (semi-)supervised classification schemes. Experimental results on two publicly available action recognition databases denote its effectiveness.",

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

year = "2015",

month = jan,

doi = "10.1109/ICPR.2014.239",

language = "English",

isbn = "9781479952106",

series = "Proceedings of the International Conference on Pattern Recognition (ICPR)",

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

pages = "1336--1341",

booktitle = "2014 22nd International Conference on Pattern Recognition (ICPR 2014)",

address = "United States",

note = "IEEE International Conference on Pattern Recognition (ICPR) ; Conference date: 24-08-2014 Through 28-08-2014",

}

Iosifidis, A, Tefas, A & Pitas, I 2015, Semi-supervised classification of human actions based on Neural Networks. in 2014 22nd International Conference on Pattern Recognition (ICPR 2014): Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden. Proceedings of the International Conference on Pattern Recognition (ICPR), Institute of Electrical and Electronics Engineers (IEEE), pp. 1336-1341, IEEE International Conference on Pattern Recognition (ICPR), Stockholm, Sweden, 24/08/14. https://doi.org/10.1109/ICPR.2014.239

Semi-supervised classification of human actions based on Neural Networks. / Iosifidis, Alexandros; Tefas, Anastasios; Pitas, Ioannis.
2014 22nd International Conference on Pattern Recognition (ICPR 2014): Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE), 2015. p. 1336-1341 (Proceedings of the International Conference on Pattern Recognition (ICPR)).

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

TY - GEN

T1 - Semi-supervised classification of human actions based on Neural Networks

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Pitas, Ioannis

PY - 2015/1

Y1 - 2015/1

N2 - In this paper, we propose a novel algorithm for Single-hidden Layer Feedforward Neural networks training which is able to exploit information coming from both labeled and unlabeled data for semi-supervised action classification. We extend the Extreme Learning Machine algorithm by incorporating appropriate regularization terms describing geometric properties and discrimination criteria of the training data representation in the ELM space to this end. The proposed algorithm is evaluated on human action recognition, where its performance is compared with that of other (semi-)supervised classification schemes. Experimental results on two publicly available action recognition databases denote its effectiveness.

AB - In this paper, we propose a novel algorithm for Single-hidden Layer Feedforward Neural networks training which is able to exploit information coming from both labeled and unlabeled data for semi-supervised action classification. We extend the Extreme Learning Machine algorithm by incorporating appropriate regularization terms describing geometric properties and discrimination criteria of the training data representation in the ELM space to this end. The proposed algorithm is evaluated on human action recognition, where its performance is compared with that of other (semi-)supervised classification schemes. Experimental results on two publicly available action recognition databases denote its effectiveness.

U2 - 10.1109/ICPR.2014.239

DO - 10.1109/ICPR.2014.239

M3 - Conference Contribution (Conference Proceeding)

SN - 9781479952106

T3 - Proceedings of the International Conference on Pattern Recognition (ICPR)

SP - 1336

EP - 1341

BT - 2014 22nd International Conference on Pattern Recognition (ICPR 2014)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE International Conference on Pattern Recognition (ICPR)

Y2 - 24 August 2014 through 28 August 2014

ER -

Iosifidis A, Tefas A, Pitas I. Semi-supervised classification of human actions based on Neural Networks. In 2014 22nd International Conference on Pattern Recognition (ICPR 2014): Proceedings of a meeting held 24-28 August 2014, Stockholm, Sweden. Institute of Electrical and Electronics Engineers (IEEE). 2015. p. 1336-1341. (Proceedings of the International Conference on Pattern Recognition (ICPR)). Epub 2014 Dec 6. doi: 10.1109/ICPR.2014.239

Semi-supervised classification of human actions based on Neural Networks

Abstract

Publication series

Conference

Access to Document

Handle.net

Fingerprint

Cite this