Detection, localization and pose classification of ear in 3D face profile images

Jiajia Lei, Jindan Zhou, Mohamed Abdel-Mottaleb, Xinge You

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

We present an efficient and robust system for landmark localization, segmentation and pose classification of ears from 3D profile facial range data. After defining 18 landmarks on the ear, including Triangular Fossa and Incisure Intertragica, a novel Ear Tree-structured Graph (ETG) is proposed to represent the 3D ear. We trained a flexible mixture model to locate these landmarks automatically. Afterwards, the ear region is outlined as the minimum rectangle including all landmarks. Finally, by calculating the turning angle between landmarks on the helix, the ear is classified as either a left or a right ear. To the best of our knowledge, there is no previous work on automatic landmark localization for 3D ear on 3D facial profile depth images. Experiments are conducted on University of Notre Dame Collection F and Collection J2 datasets, containing large occlusion, scale and pose variations. Results demonstrate the effectiveness of the proposed techniques.

Original languageEnglish
Title of host publication2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings
Pages4200-4204
Number of pages5
DOIs
StatePublished - Dec 1 2013
Event2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, Australia
Duration: Sep 15 2013Sep 18 2013

Other

Other2013 20th IEEE International Conference on Image Processing, ICIP 2013
CountryAustralia
CityMelbourne, VIC
Period9/15/139/18/13

Fingerprint

Experiments

Keywords

  • ear detection
  • ear tree-structured graph
  • flexible mixture model
  • Landmark localization
  • pose classification

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Lei, J., Zhou, J., Abdel-Mottaleb, M., & You, X. (2013). Detection, localization and pose classification of ear in 3D face profile images. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings (pp. 4200-4204). [6738865] https://doi.org/10.1109/ICIP.2013.6738865

Detection, localization and pose classification of ear in 3D face profile images. / Lei, Jiajia; Zhou, Jindan; Abdel-Mottaleb, Mohamed; You, Xinge.

2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 4200-4204 6738865.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lei, J, Zhou, J, Abdel-Mottaleb, M & You, X 2013, Detection, localization and pose classification of ear in 3D face profile images. in 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings., 6738865, pp. 4200-4204, 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, Australia, 9/15/13. https://doi.org/10.1109/ICIP.2013.6738865
Lei J, Zhou J, Abdel-Mottaleb M, You X. Detection, localization and pose classification of ear in 3D face profile images. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 4200-4204. 6738865 https://doi.org/10.1109/ICIP.2013.6738865
Lei, Jiajia ; Zhou, Jindan ; Abdel-Mottaleb, Mohamed ; You, Xinge. / Detection, localization and pose classification of ear in 3D face profile images. 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. pp. 4200-4204
@inproceedings{254699ec6e5d4276976363feefc8ec06,
title = "Detection, localization and pose classification of ear in 3D face profile images",
abstract = "We present an efficient and robust system for landmark localization, segmentation and pose classification of ears from 3D profile facial range data. After defining 18 landmarks on the ear, including Triangular Fossa and Incisure Intertragica, a novel Ear Tree-structured Graph (ETG) is proposed to represent the 3D ear. We trained a flexible mixture model to locate these landmarks automatically. Afterwards, the ear region is outlined as the minimum rectangle including all landmarks. Finally, by calculating the turning angle between landmarks on the helix, the ear is classified as either a left or a right ear. To the best of our knowledge, there is no previous work on automatic landmark localization for 3D ear on 3D facial profile depth images. Experiments are conducted on University of Notre Dame Collection F and Collection J2 datasets, containing large occlusion, scale and pose variations. Results demonstrate the effectiveness of the proposed techniques.",
keywords = "ear detection, ear tree-structured graph, flexible mixture model, Landmark localization, pose classification",
author = "Jiajia Lei and Jindan Zhou and Mohamed Abdel-Mottaleb and Xinge You",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ICIP.2013.6738865",
language = "English",
isbn = "9781479923410",
pages = "4200--4204",
booktitle = "2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings",

}

TY - GEN

T1 - Detection, localization and pose classification of ear in 3D face profile images

AU - Lei, Jiajia

AU - Zhou, Jindan

AU - Abdel-Mottaleb, Mohamed

AU - You, Xinge

PY - 2013/12/1

Y1 - 2013/12/1

N2 - We present an efficient and robust system for landmark localization, segmentation and pose classification of ears from 3D profile facial range data. After defining 18 landmarks on the ear, including Triangular Fossa and Incisure Intertragica, a novel Ear Tree-structured Graph (ETG) is proposed to represent the 3D ear. We trained a flexible mixture model to locate these landmarks automatically. Afterwards, the ear region is outlined as the minimum rectangle including all landmarks. Finally, by calculating the turning angle between landmarks on the helix, the ear is classified as either a left or a right ear. To the best of our knowledge, there is no previous work on automatic landmark localization for 3D ear on 3D facial profile depth images. Experiments are conducted on University of Notre Dame Collection F and Collection J2 datasets, containing large occlusion, scale and pose variations. Results demonstrate the effectiveness of the proposed techniques.

AB - We present an efficient and robust system for landmark localization, segmentation and pose classification of ears from 3D profile facial range data. After defining 18 landmarks on the ear, including Triangular Fossa and Incisure Intertragica, a novel Ear Tree-structured Graph (ETG) is proposed to represent the 3D ear. We trained a flexible mixture model to locate these landmarks automatically. Afterwards, the ear region is outlined as the minimum rectangle including all landmarks. Finally, by calculating the turning angle between landmarks on the helix, the ear is classified as either a left or a right ear. To the best of our knowledge, there is no previous work on automatic landmark localization for 3D ear on 3D facial profile depth images. Experiments are conducted on University of Notre Dame Collection F and Collection J2 datasets, containing large occlusion, scale and pose variations. Results demonstrate the effectiveness of the proposed techniques.

KW - ear detection

KW - ear tree-structured graph

KW - flexible mixture model

KW - Landmark localization

KW - pose classification

UR - http://www.scopus.com/inward/record.url?scp=84897766602&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897766602&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2013.6738865

DO - 10.1109/ICIP.2013.6738865

M3 - Conference contribution

AN - SCOPUS:84897766602

SN - 9781479923410

SP - 4200

EP - 4204

BT - 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings

ER -