Depth based dual component dynamic gesture recognition

Helman Stern; Kiril Smilansky; Sigal Berman

Depth based dual component dynamic gesture recognition

Helman Stern, Kiril Smilansky, Sigal Berman

Department of Industrial Engineering and Management

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

3 Scopus citations

Abstract

In this paper we describe several approaches for recognition of gestures that include simultaneous arm motion and hand configuration variations. Based on compound (dual component) gestures selected from the ASL we developed methods for recognizing such gestures from Kinect sensor videos. The method consists of hand segmentation from depth images followed by feature extraction based on block partitioning of the hand image. When combined with trajectory features a single-stage classifier is obtained. A second method which classifies arm movement and hand configuration in two-stages is also developed. These two methods are compared to a moment based classifier from the literature. Using a database of 11 subjects for training and testing the average classification accuracy of the one-stage classifier was the highest (95.5%) and that of the moment based classifier was the lowest (20.9%). The two-stage classifier obtained an average classification accuracy of61.1%.

Original language	English
Title of host publication	Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013
Editors	Hamid R. Arabnia, Leonidas Deligiannidis, Joan Lu, Fernando G. Tinetti, Jane You, George Jandieri, Gerald Schaefer, Ashu M. G. Solo, Vladimir Volkov
Publisher	CSREA Press
Pages	985-991
Number of pages	7
ISBN (Electronic)	1601322534, 9781601322531
State	Published - 1 Jan 2013
Event	2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, at WORLDCOMP 2013 - Las Vegas, United States Duration: 22 Jul 2013 → 25 Jul 2013

Publication series

Name	Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013
Volume	2

Conference

Conference	2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, at WORLDCOMP 2013
Country/Territory	United States
City	Las Vegas
Period	22/07/13 → 25/07/13

Keywords

Dynamic motion gestures
Gesture recognition
Human-machine interaction
Sign language

ASJC Scopus subject areas

Computer Graphics and Computer-Aided Design
Computer Vision and Pattern Recognition

Cite this

Stern, H., Smilansky, K., & Berman, S. (2013). Depth based dual component dynamic gesture recognition. In H. R. Arabnia, L. Deligiannidis, J. Lu, F. G. Tinetti, J. You, G. Jandieri, G. Schaefer, A. M. G. Solo, & V. Volkov (Eds.), Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013 (pp. 985-991). (Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013; Vol. 2). CSREA Press.

Stern, Helman ; Smilansky, Kiril ; Berman, Sigal. / Depth based dual component dynamic gesture recognition. Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013. editor / Hamid R. Arabnia ; Leonidas Deligiannidis ; Joan Lu ; Fernando G. Tinetti ; Jane You ; George Jandieri ; Gerald Schaefer ; Ashu M. G. Solo ; Vladimir Volkov. CSREA Press, 2013. pp. 985-991 (Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013).

@inproceedings{c16f9fb42a924da7b6f047b79f10bda2,

title = "Depth based dual component dynamic gesture recognition",

abstract = "In this paper we describe several approaches for recognition of gestures that include simultaneous arm motion and hand configuration variations. Based on compound (dual component) gestures selected from the ASL we developed methods for recognizing such gestures from Kinect sensor videos. The method consists of hand segmentation from depth images followed by feature extraction based on block partitioning of the hand image. When combined with trajectory features a single-stage classifier is obtained. A second method which classifies arm movement and hand configuration in two-stages is also developed. These two methods are compared to a moment based classifier from the literature. Using a database of 11 subjects for training and testing the average classification accuracy of the one-stage classifier was the highest (95.5%) and that of the moment based classifier was the lowest (20.9%). The two-stage classifier obtained an average classification accuracy of61.1%.",

keywords = "Dynamic motion gestures, Gesture recognition, Human-machine interaction, Sign language",

author = "Helman Stern and Kiril Smilansky and Sigal Berman",

note = "Publisher Copyright: {\textcopyright} 2013 CSREA Press. All rights reserved.; 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, at WORLDCOMP 2013 ; Conference date: 22-07-2013 Through 25-07-2013",

year = "2013",

month = jan,

day = "1",

language = "English",

series = "Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013",

publisher = "CSREA Press",

pages = "985--991",

editor = "Arabnia, {Hamid R.} and Leonidas Deligiannidis and Joan Lu and Tinetti, {Fernando G.} and Jane You and George Jandieri and Gerald Schaefer and Solo, {Ashu M. G.} and Vladimir Volkov",

booktitle = "Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013",

}

Stern, H, Smilansky, K & Berman, S 2013, Depth based dual component dynamic gesture recognition. in HR Arabnia, L Deligiannidis, J Lu, FG Tinetti, J You, G Jandieri, G Schaefer, AMG Solo & V Volkov (eds), Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013. Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, vol. 2, CSREA Press, pp. 985-991, 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, at WORLDCOMP 2013, Las Vegas, United States, 22/07/13.

Depth based dual component dynamic gesture recognition. / Stern, Helman; Smilansky, Kiril; Berman, Sigal.
Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013. ed. / Hamid R. Arabnia; Leonidas Deligiannidis; Joan Lu; Fernando G. Tinetti; Jane You; George Jandieri; Gerald Schaefer; Ashu M. G. Solo; Vladimir Volkov. CSREA Press, 2013. p. 985-991 (Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013; Vol. 2).

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

TY - GEN

T1 - Depth based dual component dynamic gesture recognition

AU - Stern, Helman

AU - Smilansky, Kiril

AU - Berman, Sigal

PY - 2013/1/1

Y1 - 2013/1/1

N2 - In this paper we describe several approaches for recognition of gestures that include simultaneous arm motion and hand configuration variations. Based on compound (dual component) gestures selected from the ASL we developed methods for recognizing such gestures from Kinect sensor videos. The method consists of hand segmentation from depth images followed by feature extraction based on block partitioning of the hand image. When combined with trajectory features a single-stage classifier is obtained. A second method which classifies arm movement and hand configuration in two-stages is also developed. These two methods are compared to a moment based classifier from the literature. Using a database of 11 subjects for training and testing the average classification accuracy of the one-stage classifier was the highest (95.5%) and that of the moment based classifier was the lowest (20.9%). The two-stage classifier obtained an average classification accuracy of61.1%.

AB - In this paper we describe several approaches for recognition of gestures that include simultaneous arm motion and hand configuration variations. Based on compound (dual component) gestures selected from the ASL we developed methods for recognizing such gestures from Kinect sensor videos. The method consists of hand segmentation from depth images followed by feature extraction based on block partitioning of the hand image. When combined with trajectory features a single-stage classifier is obtained. A second method which classifies arm movement and hand configuration in two-stages is also developed. These two methods are compared to a moment based classifier from the literature. Using a database of 11 subjects for training and testing the average classification accuracy of the one-stage classifier was the highest (95.5%) and that of the moment based classifier was the lowest (20.9%). The two-stage classifier obtained an average classification accuracy of61.1%.

KW - Dynamic motion gestures

KW - Gesture recognition

KW - Human-machine interaction

KW - Sign language

UR - http://www.scopus.com/inward/record.url?scp=85048688660&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85048688660

T3 - Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013

SP - 985

EP - 991

BT - Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013

A2 - Arabnia, Hamid R.

A2 - Deligiannidis, Leonidas

A2 - Lu, Joan

A2 - Tinetti, Fernando G.

A2 - You, Jane

A2 - Jandieri, George

A2 - Schaefer, Gerald

A2 - Solo, Ashu M. G.

A2 - Volkov, Vladimir

PB - CSREA Press

T2 - 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013, at WORLDCOMP 2013

Y2 - 22 July 2013 through 25 July 2013

ER -

Stern H, Smilansky K, Berman S. Depth based dual component dynamic gesture recognition. In Arabnia HR, Deligiannidis L, Lu J, Tinetti FG, You J, Jandieri G, Schaefer G, Solo AMG, Volkov V, editors, Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013. CSREA Press. 2013. p. 985-991. (Proceedings of the 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2013).

Depth based dual component dynamic gesture recognition

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this