3D head pose tracking using a particle filter with nose template matching

Hitoshi Kubota, Masami Takeshi, Hideo Saito

Research output: Contribution to journalArticle

Abstract

In this paper, we propose a method for real-time tracking of the driver's head pose in a real vehicle environment by using multiple NIR cameras. In order to achieve real-time performance and high accuracy, the head pose is estimated with six degrees of freedom by a particle filter with a restricted state space. First, the 3D position of the nostrils is measured by template matching of stereo images. Because the nostrils are the darkest area in the captured NTR images, we can detect the position of the nose robustly and very precisely. From the 3D position of the nose, we can make an initial estimate of the head pose. Then, each hypothesis is updated by prior probability with the constraint of the nose position. Thus, it is possible to reduce the number of particles while maintaining accuracy. Experimental results indicate that this method is effective for head pose tracking.

Original languageEnglish
Pages (from-to)34-42
Number of pages9
JournalElectronics and Communications in Japan
Volume94
Issue number1
DOIs
Publication statusPublished - 2011 Jan

Fingerprint

Template matching
Template Matching
Particle Filter
templates
Cameras
filters
Real-time
Prior Probability
Driver
High Accuracy
vehicles
State Space
degrees of freedom
Camera
Degree of freedom
cameras
Experimental Results
estimates
Estimate

Keywords

  • driver
  • head pose tracking
  • nose hole
  • particle filter
  • template matching
  • view direction tracking

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Networks and Communications
  • Physics and Astronomy(all)
  • Signal Processing
  • Applied Mathematics

Cite this

3D head pose tracking using a particle filter with nose template matching. / Kubota, Hitoshi; Takeshi, Masami; Saito, Hideo.

In: Electronics and Communications in Japan, Vol. 94, No. 1, 01.2011, p. 34-42.

Research output: Contribution to journalArticle

@article{90b3ec884c7d4b8a98d3c3e3072d5c77,
title = "3D head pose tracking using a particle filter with nose template matching",
abstract = "In this paper, we propose a method for real-time tracking of the driver's head pose in a real vehicle environment by using multiple NIR cameras. In order to achieve real-time performance and high accuracy, the head pose is estimated with six degrees of freedom by a particle filter with a restricted state space. First, the 3D position of the nostrils is measured by template matching of stereo images. Because the nostrils are the darkest area in the captured NTR images, we can detect the position of the nose robustly and very precisely. From the 3D position of the nose, we can make an initial estimate of the head pose. Then, each hypothesis is updated by prior probability with the constraint of the nose position. Thus, it is possible to reduce the number of particles while maintaining accuracy. Experimental results indicate that this method is effective for head pose tracking.",
keywords = "driver, head pose tracking, nose hole, particle filter, template matching, view direction tracking",
author = "Hitoshi Kubota and Masami Takeshi and Hideo Saito",
year = "2011",
month = "1",
doi = "10.1002/ecj.10268",
language = "English",
volume = "94",
pages = "34--42",
journal = "Electronics and Communications in Japan",
issn = "1942-9533",
publisher = "Scripta Technica",
number = "1",

}

TY - JOUR

T1 - 3D head pose tracking using a particle filter with nose template matching

AU - Kubota, Hitoshi

AU - Takeshi, Masami

AU - Saito, Hideo

PY - 2011/1

Y1 - 2011/1

N2 - In this paper, we propose a method for real-time tracking of the driver's head pose in a real vehicle environment by using multiple NIR cameras. In order to achieve real-time performance and high accuracy, the head pose is estimated with six degrees of freedom by a particle filter with a restricted state space. First, the 3D position of the nostrils is measured by template matching of stereo images. Because the nostrils are the darkest area in the captured NTR images, we can detect the position of the nose robustly and very precisely. From the 3D position of the nose, we can make an initial estimate of the head pose. Then, each hypothesis is updated by prior probability with the constraint of the nose position. Thus, it is possible to reduce the number of particles while maintaining accuracy. Experimental results indicate that this method is effective for head pose tracking.

AB - In this paper, we propose a method for real-time tracking of the driver's head pose in a real vehicle environment by using multiple NIR cameras. In order to achieve real-time performance and high accuracy, the head pose is estimated with six degrees of freedom by a particle filter with a restricted state space. First, the 3D position of the nostrils is measured by template matching of stereo images. Because the nostrils are the darkest area in the captured NTR images, we can detect the position of the nose robustly and very precisely. From the 3D position of the nose, we can make an initial estimate of the head pose. Then, each hypothesis is updated by prior probability with the constraint of the nose position. Thus, it is possible to reduce the number of particles while maintaining accuracy. Experimental results indicate that this method is effective for head pose tracking.

KW - driver

KW - head pose tracking

KW - nose hole

KW - particle filter

KW - template matching

KW - view direction tracking

UR - http://www.scopus.com/inward/record.url?scp=78651060620&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651060620&partnerID=8YFLogxK

U2 - 10.1002/ecj.10268

DO - 10.1002/ecj.10268

M3 - Article

AN - SCOPUS:78651060620

VL - 94

SP - 34

EP - 42

JO - Electronics and Communications in Japan

JF - Electronics and Communications in Japan

SN - 1942-9533

IS - 1

ER -