GesID: 3D gesture authentication based on depth camera and one-class classification

Xuan Wang, Jiro Tanaka

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Biometric authentication is popular in authentication systems, and gesture as a carrier of behavior characteristics has the advantages of being difficult to imitate and containing abundant information. This research aims to use three-dimensional (3D) depth information of gesture movement to perform authentication with less user effort. We propose an approach based on depth cameras, which satisfies three requirements: Can authenticate from a single, customized gesture; achieves high accuracy without an excessive number of gestures for training; and continues learning the gesture during use of the system. To satisfy these requirements respectively: We use a sparse autoencoder to memorize the single gesture; we employ data augmentation technology to solve the problem of insufficient data; and we use incremental learning technology for allowing the system to memorize the gesture incrementally over time. An experiment has been performed on different gestures in different user situations that demonstrates the accuracy of one-class classification (OCC), and proves the effectiveness and reliability of the approach. Gesture authentication based on 3D depth cameras could be achieved with reduced user effort.

Original languageEnglish
Article number3265
JournalSensors (Switzerland)
Volume18
Issue number10
DOIs
Publication statusPublished - 2018 Oct 1

Fingerprint

Gestures
Authentication
Cameras
cameras
learning
biometrics
requirements
education
Biometrics
augmentation
Biometric Identification
Learning
Technology
Experiments

Keywords

  • Authentication
  • Depth camera
  • Gesture
  • Incremental learning
  • Neural network
  • One-class classification
  • Sparse autoencoder

ASJC Scopus subject areas

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

GesID : 3D gesture authentication based on depth camera and one-class classification. / Wang, Xuan; Tanaka, Jiro.

In: Sensors (Switzerland), Vol. 18, No. 10, 3265, 01.10.2018.

Research output: Contribution to journalArticle

@article{1fb4134a83c34dbf8ff2c6e665a17328,
title = "GesID: 3D gesture authentication based on depth camera and one-class classification",
abstract = "Biometric authentication is popular in authentication systems, and gesture as a carrier of behavior characteristics has the advantages of being difficult to imitate and containing abundant information. This research aims to use three-dimensional (3D) depth information of gesture movement to perform authentication with less user effort. We propose an approach based on depth cameras, which satisfies three requirements: Can authenticate from a single, customized gesture; achieves high accuracy without an excessive number of gestures for training; and continues learning the gesture during use of the system. To satisfy these requirements respectively: We use a sparse autoencoder to memorize the single gesture; we employ data augmentation technology to solve the problem of insufficient data; and we use incremental learning technology for allowing the system to memorize the gesture incrementally over time. An experiment has been performed on different gestures in different user situations that demonstrates the accuracy of one-class classification (OCC), and proves the effectiveness and reliability of the approach. Gesture authentication based on 3D depth cameras could be achieved with reduced user effort.",
keywords = "Authentication, Depth camera, Gesture, Incremental learning, Neural network, One-class classification, Sparse autoencoder",
author = "Xuan Wang and Jiro Tanaka",
year = "2018",
month = "10",
day = "1",
doi = "10.3390/s18103265",
language = "English",
volume = "18",
journal = "Sensors (Switzerland)",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "10",

}

TY - JOUR

T1 - GesID

T2 - 3D gesture authentication based on depth camera and one-class classification

AU - Wang, Xuan

AU - Tanaka, Jiro

PY - 2018/10/1

Y1 - 2018/10/1

N2 - Biometric authentication is popular in authentication systems, and gesture as a carrier of behavior characteristics has the advantages of being difficult to imitate and containing abundant information. This research aims to use three-dimensional (3D) depth information of gesture movement to perform authentication with less user effort. We propose an approach based on depth cameras, which satisfies three requirements: Can authenticate from a single, customized gesture; achieves high accuracy without an excessive number of gestures for training; and continues learning the gesture during use of the system. To satisfy these requirements respectively: We use a sparse autoencoder to memorize the single gesture; we employ data augmentation technology to solve the problem of insufficient data; and we use incremental learning technology for allowing the system to memorize the gesture incrementally over time. An experiment has been performed on different gestures in different user situations that demonstrates the accuracy of one-class classification (OCC), and proves the effectiveness and reliability of the approach. Gesture authentication based on 3D depth cameras could be achieved with reduced user effort.

AB - Biometric authentication is popular in authentication systems, and gesture as a carrier of behavior characteristics has the advantages of being difficult to imitate and containing abundant information. This research aims to use three-dimensional (3D) depth information of gesture movement to perform authentication with less user effort. We propose an approach based on depth cameras, which satisfies three requirements: Can authenticate from a single, customized gesture; achieves high accuracy without an excessive number of gestures for training; and continues learning the gesture during use of the system. To satisfy these requirements respectively: We use a sparse autoencoder to memorize the single gesture; we employ data augmentation technology to solve the problem of insufficient data; and we use incremental learning technology for allowing the system to memorize the gesture incrementally over time. An experiment has been performed on different gestures in different user situations that demonstrates the accuracy of one-class classification (OCC), and proves the effectiveness and reliability of the approach. Gesture authentication based on 3D depth cameras could be achieved with reduced user effort.

KW - Authentication

KW - Depth camera

KW - Gesture

KW - Incremental learning

KW - Neural network

KW - One-class classification

KW - Sparse autoencoder

UR - http://www.scopus.com/inward/record.url?scp=85054072695&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054072695&partnerID=8YFLogxK

U2 - 10.3390/s18103265

DO - 10.3390/s18103265

M3 - Article

C2 - 30274187

AN - SCOPUS:85054072695

VL - 18

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-3210

IS - 10

M1 - 3265

ER -