Dual-camera 3D head tracking for clinical infant monitoring

R.W.J.J. Saeijs, W.E. Tjon A Ten, P.H.N. de With

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

This paper presents a new algorithm for dual-camera 3D head tracking, intended for clinical infant monitoring. The paper includes a brief motivation with reference to the state-of-the-art in face-related image analysis. The proposed algorithm uses a clipped-ellipsoid head model and 3D head pose recovery by joint alignment of paired templates based on dense-HOG features. In the algorithm, template pairs are dynamically extracted and a limited number of template pairs are stored and re-used for drift reduction. We report experimental results on real-life videos of infants in bed in a hospital, captured in visual light as well as near-infrared light. Results show consistently good tracking behavior. For challenging video sequences, the mean tracking error in terms of endocanthion location error relative to the innercanthal distance remains below 30%. This error has proven to be sufficiently low for 3D head tracking to support infant face analysis. For this reason, the proposed algorithm is used successfully in an infant monitoring system under development.

Original languageEnglish
Title of host publication2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages1-8
Number of pages8
ISBN (Electronic)978-1-5386-4396-9
ISBN (Print)978-1-5386-4397-6
DOIs
Publication statusPublished - 3 May 2018
Event2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018 - Fez, Morocco
Duration: 2 Apr 20184 Apr 2018

Conference

Conference2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018
CountryMorocco
CityFez
Period2/04/184/04/18

Fingerprint

Cameras
Monitoring
Hospital beds
Image analysis
Infrared radiation
Recovery

Keywords

  • 3D head tracking
  • dense HOG
  • dual camera
  • infant monitoring

Cite this

Saeijs, R. W. J. J., Tjon A Ten, W. E., & de With, P. H. N. (2018). Dual-camera 3D head tracking for clinical infant monitoring. In 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018 (pp. 1-8). Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ISACV.2018.8354068
Saeijs, R.W.J.J. ; Tjon A Ten, W.E. ; de With, P.H.N. / Dual-camera 3D head tracking for clinical infant monitoring. 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018. Piscataway : Institute of Electrical and Electronics Engineers, 2018. pp. 1-8
@inproceedings{c1dc80ebebfc4c3aa211ffa380c843a6,
title = "Dual-camera 3D head tracking for clinical infant monitoring",
abstract = "This paper presents a new algorithm for dual-camera 3D head tracking, intended for clinical infant monitoring. The paper includes a brief motivation with reference to the state-of-the-art in face-related image analysis. The proposed algorithm uses a clipped-ellipsoid head model and 3D head pose recovery by joint alignment of paired templates based on dense-HOG features. In the algorithm, template pairs are dynamically extracted and a limited number of template pairs are stored and re-used for drift reduction. We report experimental results on real-life videos of infants in bed in a hospital, captured in visual light as well as near-infrared light. Results show consistently good tracking behavior. For challenging video sequences, the mean tracking error in terms of endocanthion location error relative to the innercanthal distance remains below 30{\%}. This error has proven to be sufficiently low for 3D head tracking to support infant face analysis. For this reason, the proposed algorithm is used successfully in an infant monitoring system under development.",
keywords = "3D head tracking, dense HOG, dual camera, infant monitoring",
author = "R.W.J.J. Saeijs and {Tjon A Ten}, W.E. and {de With}, P.H.N.",
year = "2018",
month = "5",
day = "3",
doi = "10.1109/ISACV.2018.8354068",
language = "English",
isbn = "978-1-5386-4397-6",
pages = "1--8",
booktitle = "2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Saeijs, RWJJ, Tjon A Ten, WE & de With, PHN 2018, Dual-camera 3D head tracking for clinical infant monitoring. in 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018. Institute of Electrical and Electronics Engineers, Piscataway, pp. 1-8, 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018, Fez, Morocco, 2/04/18. https://doi.org/10.1109/ISACV.2018.8354068

Dual-camera 3D head tracking for clinical infant monitoring. / Saeijs, R.W.J.J.; Tjon A Ten, W.E.; de With, P.H.N.

2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018. Piscataway : Institute of Electrical and Electronics Engineers, 2018. p. 1-8.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Dual-camera 3D head tracking for clinical infant monitoring

AU - Saeijs, R.W.J.J.

AU - Tjon A Ten, W.E.

AU - de With, P.H.N.

PY - 2018/5/3

Y1 - 2018/5/3

N2 - This paper presents a new algorithm for dual-camera 3D head tracking, intended for clinical infant monitoring. The paper includes a brief motivation with reference to the state-of-the-art in face-related image analysis. The proposed algorithm uses a clipped-ellipsoid head model and 3D head pose recovery by joint alignment of paired templates based on dense-HOG features. In the algorithm, template pairs are dynamically extracted and a limited number of template pairs are stored and re-used for drift reduction. We report experimental results on real-life videos of infants in bed in a hospital, captured in visual light as well as near-infrared light. Results show consistently good tracking behavior. For challenging video sequences, the mean tracking error in terms of endocanthion location error relative to the innercanthal distance remains below 30%. This error has proven to be sufficiently low for 3D head tracking to support infant face analysis. For this reason, the proposed algorithm is used successfully in an infant monitoring system under development.

AB - This paper presents a new algorithm for dual-camera 3D head tracking, intended for clinical infant monitoring. The paper includes a brief motivation with reference to the state-of-the-art in face-related image analysis. The proposed algorithm uses a clipped-ellipsoid head model and 3D head pose recovery by joint alignment of paired templates based on dense-HOG features. In the algorithm, template pairs are dynamically extracted and a limited number of template pairs are stored and re-used for drift reduction. We report experimental results on real-life videos of infants in bed in a hospital, captured in visual light as well as near-infrared light. Results show consistently good tracking behavior. For challenging video sequences, the mean tracking error in terms of endocanthion location error relative to the innercanthal distance remains below 30%. This error has proven to be sufficiently low for 3D head tracking to support infant face analysis. For this reason, the proposed algorithm is used successfully in an infant monitoring system under development.

KW - 3D head tracking

KW - dense HOG

KW - dual camera

KW - infant monitoring

UR - http://www.scopus.com/inward/record.url?scp=85050939367&partnerID=8YFLogxK

U2 - 10.1109/ISACV.2018.8354068

DO - 10.1109/ISACV.2018.8354068

M3 - Conference contribution

AN - SCOPUS:85050939367

SN - 978-1-5386-4397-6

SP - 1

EP - 8

BT - 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Saeijs RWJJ, Tjon A Ten WE, de With PHN. Dual-camera 3D head tracking for clinical infant monitoring. In 2018 International Conference on Intelligent Systems and Computer Vision, ISCV 2018. Piscataway: Institute of Electrical and Electronics Engineers. 2018. p. 1-8 https://doi.org/10.1109/ISACV.2018.8354068