Gait-based person identification robust to changes in appearance

Yumi Iwashita, Koji Uchino, Ryo Kurazume

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

The identification of a person from gait images is generally sensitive to appearance changes, such as variations of clothes and belongings. One possibility to deal with this problem is to collect possible subjects' appearance changes in a database. However, it is almost impossible to predict all appearance changes in advance. In this paper, we propose a novel method, which allows robustly identifying people in spite of changes in appearance, without using a database of predicted appearance changes. In the proposed method, firstly, the human body image is divided into multiple areas, and features for each area are extracted. Next, a matching weight for each area is estimated based on the similarity between the extracted features and those in the database for standard clothes. Finally, the subject is identified by weighted integration of similarities in all areas. Experiments using the gait database CASIA show the best correct classification rate compared with conventional methods experiments.

Original languageEnglish
Pages (from-to)7884-7901
Number of pages18
JournalSensors (Switzerland)
Volume13
Issue number6
DOIs
Publication statusPublished - Jun 1 2013

Fingerprint

gait
Gait
Databases
Clothing
Body Image
human body
Experiments
Weights and Measures
Identification (Psychology)

All Science Journal Classification (ASJC) codes

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

Cite this

Gait-based person identification robust to changes in appearance. / Iwashita, Yumi; Uchino, Koji; Kurazume, Ryo.

In: Sensors (Switzerland), Vol. 13, No. 6, 01.06.2013, p. 7884-7901.

Research output: Contribution to journalArticle

Iwashita, Yumi ; Uchino, Koji ; Kurazume, Ryo. / Gait-based person identification robust to changes in appearance. In: Sensors (Switzerland). 2013 ; Vol. 13, No. 6. pp. 7884-7901.
@article{3b25845901ce44d284443640947409da,
title = "Gait-based person identification robust to changes in appearance",
abstract = "The identification of a person from gait images is generally sensitive to appearance changes, such as variations of clothes and belongings. One possibility to deal with this problem is to collect possible subjects' appearance changes in a database. However, it is almost impossible to predict all appearance changes in advance. In this paper, we propose a novel method, which allows robustly identifying people in spite of changes in appearance, without using a database of predicted appearance changes. In the proposed method, firstly, the human body image is divided into multiple areas, and features for each area are extracted. Next, a matching weight for each area is estimated based on the similarity between the extracted features and those in the database for standard clothes. Finally, the subject is identified by weighted integration of similarities in all areas. Experiments using the gait database CASIA show the best correct classification rate compared with conventional methods experiments.",
author = "Yumi Iwashita and Koji Uchino and Ryo Kurazume",
year = "2013",
month = "6",
day = "1",
doi = "10.3390/s130607884",
language = "English",
volume = "13",
pages = "7884--7901",
journal = "Sensors",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "6",

}

TY - JOUR

T1 - Gait-based person identification robust to changes in appearance

AU - Iwashita, Yumi

AU - Uchino, Koji

AU - Kurazume, Ryo

PY - 2013/6/1

Y1 - 2013/6/1

N2 - The identification of a person from gait images is generally sensitive to appearance changes, such as variations of clothes and belongings. One possibility to deal with this problem is to collect possible subjects' appearance changes in a database. However, it is almost impossible to predict all appearance changes in advance. In this paper, we propose a novel method, which allows robustly identifying people in spite of changes in appearance, without using a database of predicted appearance changes. In the proposed method, firstly, the human body image is divided into multiple areas, and features for each area are extracted. Next, a matching weight for each area is estimated based on the similarity between the extracted features and those in the database for standard clothes. Finally, the subject is identified by weighted integration of similarities in all areas. Experiments using the gait database CASIA show the best correct classification rate compared with conventional methods experiments.

AB - The identification of a person from gait images is generally sensitive to appearance changes, such as variations of clothes and belongings. One possibility to deal with this problem is to collect possible subjects' appearance changes in a database. However, it is almost impossible to predict all appearance changes in advance. In this paper, we propose a novel method, which allows robustly identifying people in spite of changes in appearance, without using a database of predicted appearance changes. In the proposed method, firstly, the human body image is divided into multiple areas, and features for each area are extracted. Next, a matching weight for each area is estimated based on the similarity between the extracted features and those in the database for standard clothes. Finally, the subject is identified by weighted integration of similarities in all areas. Experiments using the gait database CASIA show the best correct classification rate compared with conventional methods experiments.

UR - http://www.scopus.com/inward/record.url?scp=84879326287&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879326287&partnerID=8YFLogxK

U2 - 10.3390/s130607884

DO - 10.3390/s130607884

M3 - Article

C2 - 23783739

AN - SCOPUS:84879326287

VL - 13

SP - 7884

EP - 7901

JO - Sensors

JF - Sensors

SN - 1424-3210

IS - 6

ER -