Hand posture recognition using adaboost with SIFT for human robot interaction

Chieh Chih Wang; Ko Chih Wang

doi:10.1007/978-3-540-76729-9_25

Hand posture recognition using adaboost with SIFT for human robot interaction

Chieh Chih Wang^*, Ko Chih Wang

^*Corresponding author for this work

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

62 Citations (Scopus)

Abstract

Hand posture understanding is essential to human robot interaction. The existing hand detection approaches using a Viola-Jones detector have two fundamental issues, the degraded performance due to background noise in training images and the in-plane rotation variant detection. In this paper, a hand posture recognition system using the discrete Adaboost learning algorithm with Lowe's scale invariant feature transform (SIFT) features is proposed to tackle these issues simultaneously. In addition, we apply a sharing feature concept to increase the accuracy of multi-class hand posture recognition. The experimental results demonstrate that the proposed approach successfully recognizes three hand posture classes and can deal with the background noise issues. Our detector is in-plane rotation invariant, and achieves satisfactory multi-view hand detection.

Original language	English
Title of host publication	Recent Progress in Robotics
Subtitle of host publication	Viable Robotic Service to Human
Editors	Sukhan Lee, Il Hong Suh, Kim Mun Sang
Pages	317-329
Number of pages	13
DOIs	https://doi.org/10.1007/978-3-540-76729-9_25
Publication status	Published - 2008
Externally published	Yes

Publication series

Name	Lecture Notes in Control and Information Sciences
Volume	370
ISSN (Print)	0170-8643

ASJC Scopus subject areas

Library and Information Sciences

Access to Document

10.1007/978-3-540-76729-9_25

Cite this

Hand posture recognition using adaboost with SIFT for human robot interaction. / Wang, Chieh Chih; Wang, Ko Chih.
Recent Progress in Robotics: Viable Robotic Service to Human. ed. / Sukhan Lee; Il Hong Suh; Kim Mun Sang. 2008. p. 317-329 (Lecture Notes in Control and Information Sciences; Vol. 370).

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

@inproceedings{10a5842149204a1aa1d1ee5c6cc23b1f,

title = "Hand posture recognition using adaboost with SIFT for human robot interaction",

abstract = "Hand posture understanding is essential to human robot interaction. The existing hand detection approaches using a Viola-Jones detector have two fundamental issues, the degraded performance due to background noise in training images and the in-plane rotation variant detection. In this paper, a hand posture recognition system using the discrete Adaboost learning algorithm with Lowe's scale invariant feature transform (SIFT) features is proposed to tackle these issues simultaneously. In addition, we apply a sharing feature concept to increase the accuracy of multi-class hand posture recognition. The experimental results demonstrate that the proposed approach successfully recognizes three hand posture classes and can deal with the background noise issues. Our detector is in-plane rotation invariant, and achieves satisfactory multi-view hand detection.",

author = "Wang, {Chieh Chih} and Wang, {Ko Chih}",

note = "Funding Information: We acknowledge the helpful suggestions by an anonymous reviewer. This work was partially supported by grants from Taiwan NSC (#95-2218-E-002-039, #95-2221-E-002-433); Excellent Research Projects of National Taiwan University (#95R0062-AE00-05); Taiwan DOIT TDPA Program (#95-EC-17-A-04-S1-054); and Intel.",

year = "2008",

doi = "10.1007/978-3-540-76729-9_25",

language = "English",

isbn = "9783540767282",

series = "Lecture Notes in Control and Information Sciences",

pages = "317--329",

editor = "Sukhan Lee and Suh, {Il Hong} and {Mun Sang}, Kim",

booktitle = "Recent Progress in Robotics",

}

TY - GEN

T1 - Hand posture recognition using adaboost with SIFT for human robot interaction

AU - Wang, Chieh Chih

AU - Wang, Ko Chih

N1 - Funding Information: We acknowledge the helpful suggestions by an anonymous reviewer. This work was partially supported by grants from Taiwan NSC (#95-2218-E-002-039, #95-2221-E-002-433); Excellent Research Projects of National Taiwan University (#95R0062-AE00-05); Taiwan DOIT TDPA Program (#95-EC-17-A-04-S1-054); and Intel.

PY - 2008

Y1 - 2008

N2 - Hand posture understanding is essential to human robot interaction. The existing hand detection approaches using a Viola-Jones detector have two fundamental issues, the degraded performance due to background noise in training images and the in-plane rotation variant detection. In this paper, a hand posture recognition system using the discrete Adaboost learning algorithm with Lowe's scale invariant feature transform (SIFT) features is proposed to tackle these issues simultaneously. In addition, we apply a sharing feature concept to increase the accuracy of multi-class hand posture recognition. The experimental results demonstrate that the proposed approach successfully recognizes three hand posture classes and can deal with the background noise issues. Our detector is in-plane rotation invariant, and achieves satisfactory multi-view hand detection.

AB - Hand posture understanding is essential to human robot interaction. The existing hand detection approaches using a Viola-Jones detector have two fundamental issues, the degraded performance due to background noise in training images and the in-plane rotation variant detection. In this paper, a hand posture recognition system using the discrete Adaboost learning algorithm with Lowe's scale invariant feature transform (SIFT) features is proposed to tackle these issues simultaneously. In addition, we apply a sharing feature concept to increase the accuracy of multi-class hand posture recognition. The experimental results demonstrate that the proposed approach successfully recognizes three hand posture classes and can deal with the background noise issues. Our detector is in-plane rotation invariant, and achieves satisfactory multi-view hand detection.

UR - http://www.scopus.com/inward/record.url?scp=36749088130&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36749088130&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-76729-9_25

DO - 10.1007/978-3-540-76729-9_25

M3 - Conference contribution

AN - SCOPUS:36749088130

SN - 9783540767282

T3 - Lecture Notes in Control and Information Sciences

SP - 317

EP - 329

BT - Recent Progress in Robotics

A2 - Lee, Sukhan

A2 - Suh, Il Hong

A2 - Mun Sang, Kim

ER -

Hand posture recognition using adaboost with SIFT for human robot interaction

Abstract

Publication series

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this