“Dynamically Visual Learning for Person Identification with Sparsely Distributed Multiple Cameras”

Hidenori Tanaka, Hideo Saito, Itaru Kitahara, Hiroshi Murase, Kiyoshi Kogure, Norihiro Hagita

Research output: Contribution to journalArticle

Abstract

This paper proposes a dynamical visual learning method, which aims to identify person using multiple surveillance cameras sparsely distributed in space. In the proposed method, densely distributed multiple images are captured by interpolating the object' s appearance in the sparsely distributed multiple images with a simple 3D face model, and generate two initial eigenspaces (an eigenspace for pose estimation and that for identification). In case another image is captured, the object's pose and name are estimated using the eigenspaces. The image is projected onto the 3D face model as texture information to improve the object's appearance, and the eigenspaces are regenerated. The discernment capability for person identification of the proposed method is shown by experimental results.

Original languageEnglish
Pages (from-to)623-633
Number of pages11
JournalJournal of the Institute of Image Electronics Engineers of Japan
Volume38
Issue number5
DOIs
Publication statusPublished - 2009

Fingerprint

Cameras
Textures

Keywords

  • 3D face model
  • person identification
  • surveillance cameras
  • view interpolation
  • visual learning

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Electrical and Electronic Engineering

Cite this

“Dynamically Visual Learning for Person Identification with Sparsely Distributed Multiple Cameras”. / Tanaka, Hidenori; Saito, Hideo; Kitahara, Itaru; Murase, Hiroshi; Kogure, Kiyoshi; Hagita, Norihiro.

In: Journal of the Institute of Image Electronics Engineers of Japan, Vol. 38, No. 5, 2009, p. 623-633.

Research output: Contribution to journalArticle

Tanaka, Hidenori ; Saito, Hideo ; Kitahara, Itaru ; Murase, Hiroshi ; Kogure, Kiyoshi ; Hagita, Norihiro. / “Dynamically Visual Learning for Person Identification with Sparsely Distributed Multiple Cameras”. In: Journal of the Institute of Image Electronics Engineers of Japan. 2009 ; Vol. 38, No. 5. pp. 623-633.
@article{9fffd37b5f5d4b98973defc0d16583ab,
title = "“Dynamically Visual Learning for Person Identification with Sparsely Distributed Multiple Cameras”",
abstract = "This paper proposes a dynamical visual learning method, which aims to identify person using multiple surveillance cameras sparsely distributed in space. In the proposed method, densely distributed multiple images are captured by interpolating the object' s appearance in the sparsely distributed multiple images with a simple 3D face model, and generate two initial eigenspaces (an eigenspace for pose estimation and that for identification). In case another image is captured, the object's pose and name are estimated using the eigenspaces. The image is projected onto the 3D face model as texture information to improve the object's appearance, and the eigenspaces are regenerated. The discernment capability for person identification of the proposed method is shown by experimental results.",
keywords = "3D face model, person identification, surveillance cameras, view interpolation, visual learning",
author = "Hidenori Tanaka and Hideo Saito and Itaru Kitahara and Hiroshi Murase and Kiyoshi Kogure and Norihiro Hagita",
year = "2009",
doi = "10.11371/iieej.38.623",
language = "English",
volume = "38",
pages = "623--633",
journal = "Journal of the Institute of Image Electronics Engineers of Japan",
issn = "0285-9831",
publisher = "Institute of Image Electronics Engineers of Japan",
number = "5",

}

TY - JOUR

T1 - “Dynamically Visual Learning for Person Identification with Sparsely Distributed Multiple Cameras”

AU - Tanaka, Hidenori

AU - Saito, Hideo

AU - Kitahara, Itaru

AU - Murase, Hiroshi

AU - Kogure, Kiyoshi

AU - Hagita, Norihiro

PY - 2009

Y1 - 2009

N2 - This paper proposes a dynamical visual learning method, which aims to identify person using multiple surveillance cameras sparsely distributed in space. In the proposed method, densely distributed multiple images are captured by interpolating the object' s appearance in the sparsely distributed multiple images with a simple 3D face model, and generate two initial eigenspaces (an eigenspace for pose estimation and that for identification). In case another image is captured, the object's pose and name are estimated using the eigenspaces. The image is projected onto the 3D face model as texture information to improve the object's appearance, and the eigenspaces are regenerated. The discernment capability for person identification of the proposed method is shown by experimental results.

AB - This paper proposes a dynamical visual learning method, which aims to identify person using multiple surveillance cameras sparsely distributed in space. In the proposed method, densely distributed multiple images are captured by interpolating the object' s appearance in the sparsely distributed multiple images with a simple 3D face model, and generate two initial eigenspaces (an eigenspace for pose estimation and that for identification). In case another image is captured, the object's pose and name are estimated using the eigenspaces. The image is projected onto the 3D face model as texture information to improve the object's appearance, and the eigenspaces are regenerated. The discernment capability for person identification of the proposed method is shown by experimental results.

KW - 3D face model

KW - person identification

KW - surveillance cameras

KW - view interpolation

KW - visual learning

UR - http://www.scopus.com/inward/record.url?scp=85024719878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024719878&partnerID=8YFLogxK

U2 - 10.11371/iieej.38.623

DO - 10.11371/iieej.38.623

M3 - Article

VL - 38

SP - 623

EP - 633

JO - Journal of the Institute of Image Electronics Engineers of Japan

JF - Journal of the Institute of Image Electronics Engineers of Japan

SN - 0285-9831

IS - 5

ER -