Human hand tracking from binocular image sequences

Kenichi Nirei; Hideo Saito; Masaaki Mochimaru; Shinji Ozawa

Human hand tracking from binocular image sequences

Kenichi Nirei, Hideo Saito, Masaaki Mochimaru, Shinji Ozawa

Department of Information and Computer Science

Research output: Contribution to conference › Paper › peer-review

19 Citations (Scopus)

Abstract

Sensing of human hand motion is very important for a variety of applications, such as CG animation and athletic performance measurement. Tracking a hand is difficult because the hand has high degree of freedom articulated mechanisms. This paper presents a 3-D model-based hand tracking method which is robust to occlusions and local minima. Tracking is performed minimizing estimation error of an optical flow and maximizing the overlap between a projected model and a silhouette image. We employ stochastic optimization to solve them, which are generally difficult. We present experimental results on tracking from synthetic and real image sequences.

Original language	English
Pages	297-302
Number of pages	6
Publication status	Published - 1996 Dec 1
Event	Proceedings of the 1996 IEEE 22nd International Conference on Industrial Electronics, Control, and Instrumentation, IECON. Part 3 (of 3) - Taipei, Taiwan Duration: 1996 Aug 5 → 1996 Aug 10

Other

Other	Proceedings of the 1996 IEEE 22nd International Conference on Industrial Electronics, Control, and Instrumentation, IECON. Part 3 (of 3)
City	Taipei, Taiwan
Period	96/8/5 → 96/8/10

ASJC Scopus subject areas

Control and Systems Engineering
Electrical and Electronic Engineering

Cite this

@conference{b15a165e4f7f47a089b191a41980d290,

title = "Human hand tracking from binocular image sequences",

abstract = "Sensing of human hand motion is very important for a variety of applications, such as CG animation and athletic performance measurement. Tracking a hand is difficult because the hand has high degree of freedom articulated mechanisms. This paper presents a 3-D model-based hand tracking method which is robust to occlusions and local minima. Tracking is performed minimizing estimation error of an optical flow and maximizing the overlap between a projected model and a silhouette image. We employ stochastic optimization to solve them, which are generally difficult. We present experimental results on tracking from synthetic and real image sequences.",

author = "Kenichi Nirei and Hideo Saito and Masaaki Mochimaru and Shinji Ozawa",

year = "1996",

month = dec,

day = "1",

language = "English",

pages = "297--302",

note = "Proceedings of the 1996 IEEE 22nd International Conference on Industrial Electronics, Control, and Instrumentation, IECON. Part 3 (of 3) ; Conference date: 05-08-1996 Through 10-08-1996",

}

TY - CONF

T1 - Human hand tracking from binocular image sequences

AU - Nirei, Kenichi

AU - Saito, Hideo

AU - Mochimaru, Masaaki

AU - Ozawa, Shinji

PY - 1996/12/1

Y1 - 1996/12/1

N2 - Sensing of human hand motion is very important for a variety of applications, such as CG animation and athletic performance measurement. Tracking a hand is difficult because the hand has high degree of freedom articulated mechanisms. This paper presents a 3-D model-based hand tracking method which is robust to occlusions and local minima. Tracking is performed minimizing estimation error of an optical flow and maximizing the overlap between a projected model and a silhouette image. We employ stochastic optimization to solve them, which are generally difficult. We present experimental results on tracking from synthetic and real image sequences.

AB - Sensing of human hand motion is very important for a variety of applications, such as CG animation and athletic performance measurement. Tracking a hand is difficult because the hand has high degree of freedom articulated mechanisms. This paper presents a 3-D model-based hand tracking method which is robust to occlusions and local minima. Tracking is performed minimizing estimation error of an optical flow and maximizing the overlap between a projected model and a silhouette image. We employ stochastic optimization to solve them, which are generally difficult. We present experimental results on tracking from synthetic and real image sequences.

UR - http://www.scopus.com/inward/record.url?scp=0030396470&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030396470&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030396470

SP - 297

EP - 302

T2 - Proceedings of the 1996 IEEE 22nd International Conference on Industrial Electronics, Control, and Instrumentation, IECON. Part 3 (of 3)

Y2 - 5 August 1996 through 10 August 1996

ER -

Human hand tracking from binocular image sequences

Abstract

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this