Direct manipulation interface using multiple cameras for hand gesture recognition

Akira Utsumi, Jun Ohya

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

We propose a method to detect hand position, posture and shapes from multiple-viewpoint images. We employ a simple elliptic model and a small number of reliable image features detected in multiple-viewpoint images to estimate the pose (position and normal axis) of a human hand, where feature extraction is employed based on distance transformation. The COG (center of gravity) position and its distance value are extracted in the process. These features are robust against changes in hand shape and can produce stable pose estimations. A `best view' is selected from the estimation results, and hand shape recognition is performed based on a Fourier descriptor. This viewpoint selection approach can overcome the problem of self-occlusion. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.

Original languageEnglish
Title of host publicationProceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet
Editors Anon
Place of PublicationLos Alamitos, CA, United States
PublisherIEEE Comp Soc
Pages264-267
Number of pages4
Publication statusPublished - 1998
Externally publishedYes
EventProceedings of the 1998 International Conference on Multimedia Computing and Systems - Austin, TX, USA
Duration: 1998 Jun 281998 Jul 1

Other

OtherProceedings of the 1998 International Conference on Multimedia Computing and Systems
CityAustin, TX, USA
Period98/6/2898/7/1

Fingerprint

Gesture recognition
Cameras
Virtual reality
User interfaces
Feature extraction
Gravitation

ASJC Scopus subject areas

  • Computer Science(all)
  • Engineering(all)

Cite this

Utsumi, A., & Ohya, J. (1998). Direct manipulation interface using multiple cameras for hand gesture recognition. In Anon (Ed.), Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet (pp. 264-267). Los Alamitos, CA, United States: IEEE Comp Soc.

Direct manipulation interface using multiple cameras for hand gesture recognition. / Utsumi, Akira; Ohya, Jun.

Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet. ed. / Anon. Los Alamitos, CA, United States : IEEE Comp Soc, 1998. p. 264-267.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Utsumi, A & Ohya, J 1998, Direct manipulation interface using multiple cameras for hand gesture recognition. in Anon (ed.), Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet. IEEE Comp Soc, Los Alamitos, CA, United States, pp. 264-267, Proceedings of the 1998 International Conference on Multimedia Computing and Systems, Austin, TX, USA, 98/6/28.
Utsumi A, Ohya J. Direct manipulation interface using multiple cameras for hand gesture recognition. In Anon, editor, Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet. Los Alamitos, CA, United States: IEEE Comp Soc. 1998. p. 264-267
Utsumi, Akira ; Ohya, Jun. / Direct manipulation interface using multiple cameras for hand gesture recognition. Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet. editor / Anon. Los Alamitos, CA, United States : IEEE Comp Soc, 1998. pp. 264-267
@inproceedings{f8f30640295646c1959cb9358117d333,
title = "Direct manipulation interface using multiple cameras for hand gesture recognition",
abstract = "We propose a method to detect hand position, posture and shapes from multiple-viewpoint images. We employ a simple elliptic model and a small number of reliable image features detected in multiple-viewpoint images to estimate the pose (position and normal axis) of a human hand, where feature extraction is employed based on distance transformation. The COG (center of gravity) position and its distance value are extracted in the process. These features are robust against changes in hand shape and can produce stable pose estimations. A `best view' is selected from the estimation results, and hand shape recognition is performed based on a Fourier descriptor. This viewpoint selection approach can overcome the problem of self-occlusion. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.",
author = "Akira Utsumi and Jun Ohya",
year = "1998",
language = "English",
pages = "264--267",
editor = "Anon",
booktitle = "Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet",
publisher = "IEEE Comp Soc",

}

TY - GEN

T1 - Direct manipulation interface using multiple cameras for hand gesture recognition

AU - Utsumi, Akira

AU - Ohya, Jun

PY - 1998

Y1 - 1998

N2 - We propose a method to detect hand position, posture and shapes from multiple-viewpoint images. We employ a simple elliptic model and a small number of reliable image features detected in multiple-viewpoint images to estimate the pose (position and normal axis) of a human hand, where feature extraction is employed based on distance transformation. The COG (center of gravity) position and its distance value are extracted in the process. These features are robust against changes in hand shape and can produce stable pose estimations. A `best view' is selected from the estimation results, and hand shape recognition is performed based on a Fourier descriptor. This viewpoint selection approach can overcome the problem of self-occlusion. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.

AB - We propose a method to detect hand position, posture and shapes from multiple-viewpoint images. We employ a simple elliptic model and a small number of reliable image features detected in multiple-viewpoint images to estimate the pose (position and normal axis) of a human hand, where feature extraction is employed based on distance transformation. The COG (center of gravity) position and its distance value are extracted in the process. These features are robust against changes in hand shape and can produce stable pose estimations. A `best view' is selected from the estimation results, and hand shape recognition is performed based on a Fourier descriptor. This viewpoint selection approach can overcome the problem of self-occlusion. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.

UR - http://www.scopus.com/inward/record.url?scp=0031643724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031643724&partnerID=8YFLogxK

M3 - Conference contribution

SP - 264

EP - 267

BT - Proceedings of the IEEE Conference on Protocols for Multimedia Systems and Multimedia Networking, PROMS-MmNet

A2 - Anon, null

PB - IEEE Comp Soc

CY - Los Alamitos, CA, United States

ER -