Incremental new actions learning system with limited cost and storage

Che Wei Chang; Liang Gee Chen

doi:10.1109/ISCE.2015.7177847

Incremental new actions learning system with limited cost and storage

Che Wei Chang, Liang Gee Chen

National Taiwan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In the future robotic applications, robot requires the ability not only to recognize human actions but also to learn incrementally and quickly. Therefore, we proposed an incremental action learning system for this future requirement. The proposed system can continuously learn new actions quickly with robust performance and less effort.

Original language	English
Title of host publication	2015 International Symposium on Consumer Electronics, ISCE 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781467373654
DOIs	https://doi.org/10.1109/ISCE.2015.7177847
State	Published - 04 08 2015
Externally published	Yes
Event	IEEE International Symposium on Consumer Electronics, ISCE 2015 - Madrid, Spain Duration: 24 06 2015 → 26 06 2015

Publication series

Name	Proceedings of the International Symposium on Consumer Electronics, ISCE
Volume	2015-August

Conference

Conference	IEEE International Symposium on Consumer Electronics, ISCE 2015
Country/Territory	Spain
City	Madrid
Period	24/06/15 → 26/06/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

Action recognition
Incremental learning
Nearest class mean classification
new class learning
representative data selection

Access to Document

10.1109/ISCE.2015.7177847

Cite this

Chang, C. W., & Chen, L. G. (2015). Incremental new actions learning system with limited cost and storage. In 2015 International Symposium on Consumer Electronics, ISCE 2015 Article 7177847 (Proceedings of the International Symposium on Consumer Electronics, ISCE; Vol. 2015-August). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISCE.2015.7177847

@inproceedings{1edf2624f0154c3282c4ba1f7df0c254,

title = "Incremental new actions learning system with limited cost and storage",

abstract = "In the future robotic applications, robot requires the ability not only to recognize human actions but also to learn incrementally and quickly. Therefore, we proposed an incremental action learning system for this future requirement. The proposed system can continuously learn new actions quickly with robust performance and less effort.",

keywords = "Action recognition, Incremental learning, Nearest class mean classification, new class learning, representative data selection",

author = "Chang, \{Che Wei\} and Chen, \{Liang Gee\}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE International Symposium on Consumer Electronics, ISCE 2015 ; Conference date: 24-06-2015 Through 26-06-2015",

year = "2015",

month = aug,

day = "4",

doi = "10.1109/ISCE.2015.7177847",

language = "英语",

series = "Proceedings of the International Symposium on Consumer Electronics, ISCE",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2015 International Symposium on Consumer Electronics, ISCE 2015",

address = "美国",

}

Chang, CW & Chen, LG 2015, Incremental new actions learning system with limited cost and storage. in 2015 International Symposium on Consumer Electronics, ISCE 2015., 7177847, Proceedings of the International Symposium on Consumer Electronics, ISCE, vol. 2015-August, Institute of Electrical and Electronics Engineers Inc., IEEE International Symposium on Consumer Electronics, ISCE 2015, Madrid, Spain, 24/06/15. https://doi.org/10.1109/ISCE.2015.7177847

Incremental new actions learning system with limited cost and storage. / Chang, Che Wei; Chen, Liang Gee.
2015 International Symposium on Consumer Electronics, ISCE 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7177847 (Proceedings of the International Symposium on Consumer Electronics, ISCE; Vol. 2015-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Incremental new actions learning system with limited cost and storage

AU - Chang, Che Wei

AU - Chen, Liang Gee

PY - 2015/8/4

Y1 - 2015/8/4

N2 - In the future robotic applications, robot requires the ability not only to recognize human actions but also to learn incrementally and quickly. Therefore, we proposed an incremental action learning system for this future requirement. The proposed system can continuously learn new actions quickly with robust performance and less effort.

AB - In the future robotic applications, robot requires the ability not only to recognize human actions but also to learn incrementally and quickly. Therefore, we proposed an incremental action learning system for this future requirement. The proposed system can continuously learn new actions quickly with robust performance and less effort.

KW - Action recognition

KW - Incremental learning

KW - Nearest class mean classification

KW - new class learning

KW - representative data selection

UR - https://www.scopus.com/pages/publications/84953331469

U2 - 10.1109/ISCE.2015.7177847

DO - 10.1109/ISCE.2015.7177847

M3 - 会议稿件

AN - SCOPUS:84953331469

T3 - Proceedings of the International Symposium on Consumer Electronics, ISCE

BT - 2015 International Symposium on Consumer Electronics, ISCE 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE International Symposium on Consumer Electronics, ISCE 2015

Y2 - 24 June 2015 through 26 June 2015

ER -

Incremental new actions learning system with limited cost and storage

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this