Human identification based on 3D ear models

Steven Cadavid, Mohamed Abdel-Mottaleb

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

12 Citations (Scopus)

Abstract

Two 3D ear recognition systems using structure from motion (SFM) and shape from shading (SFS) techniques, respectively, are explored. Segmentation of the ear region is performed using interpolation of ridges and ravines identified in each frame in a video sequence. For the SFM system, salient features are tracked across the video sequence and are reconstructed in 3D using a factorization method. Reconstructed points located within the valid ear region are stored as the ear model. The dataset used consists of video sequences for 48 subjects. Each test model is optimally aligned to the database models using a combination of geometric transformations which result in a minimal partial Hausdorff distance. For the SFS system, the ear structure is recovered by using reflectance and illumination properties of the scene. Shape matching is performed via iterative closest point. Based on our results, we conclude that both structure from motion and shape from shading are viable approaches for 3D ear recognition from video sequences.

Original languageEnglish
Title of host publicationIEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07
DOIs
StatePublished - Dec 1 2007
Event1st IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS '07 - Crystal City, VA, United States
Duration: Sep 27 2007Sep 29 2007

Other

Other1st IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS '07
CountryUnited States
CityCrystal City, VA
Period9/27/079/29/07

Fingerprint

Identification (control systems)
Factorization
Interpolation
Lighting

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Cadavid, S., & Abdel-Mottaleb, M. (2007). Human identification based on 3D ear models. In IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07 [4401938] https://doi.org/10.1109/BTAS.2007.4401938

Human identification based on 3D ear models. / Cadavid, Steven; Abdel-Mottaleb, Mohamed.

IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07. 2007. 4401938.

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

Cadavid, S & Abdel-Mottaleb, M 2007, Human identification based on 3D ear models. in IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07., 4401938, 1st IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS '07, Crystal City, VA, United States, 9/27/07. https://doi.org/10.1109/BTAS.2007.4401938
Cadavid S, Abdel-Mottaleb M. Human identification based on 3D ear models. In IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07. 2007. 4401938 https://doi.org/10.1109/BTAS.2007.4401938
Cadavid, Steven ; Abdel-Mottaleb, Mohamed. / Human identification based on 3D ear models. IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07. 2007.
@inproceedings{2264e040638c43f6924f9abb32e64c20,
title = "Human identification based on 3D ear models",
abstract = "Two 3D ear recognition systems using structure from motion (SFM) and shape from shading (SFS) techniques, respectively, are explored. Segmentation of the ear region is performed using interpolation of ridges and ravines identified in each frame in a video sequence. For the SFM system, salient features are tracked across the video sequence and are reconstructed in 3D using a factorization method. Reconstructed points located within the valid ear region are stored as the ear model. The dataset used consists of video sequences for 48 subjects. Each test model is optimally aligned to the database models using a combination of geometric transformations which result in a minimal partial Hausdorff distance. For the SFS system, the ear structure is recovered by using reflectance and illumination properties of the scene. Shape matching is performed via iterative closest point. Based on our results, we conclude that both structure from motion and shape from shading are viable approaches for 3D ear recognition from video sequences.",
author = "Steven Cadavid and Mohamed Abdel-Mottaleb",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/BTAS.2007.4401938",
language = "English",
isbn = "9781424415977",
booktitle = "IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07",

}

TY - GEN

T1 - Human identification based on 3D ear models

AU - Cadavid, Steven

AU - Abdel-Mottaleb, Mohamed

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Two 3D ear recognition systems using structure from motion (SFM) and shape from shading (SFS) techniques, respectively, are explored. Segmentation of the ear region is performed using interpolation of ridges and ravines identified in each frame in a video sequence. For the SFM system, salient features are tracked across the video sequence and are reconstructed in 3D using a factorization method. Reconstructed points located within the valid ear region are stored as the ear model. The dataset used consists of video sequences for 48 subjects. Each test model is optimally aligned to the database models using a combination of geometric transformations which result in a minimal partial Hausdorff distance. For the SFS system, the ear structure is recovered by using reflectance and illumination properties of the scene. Shape matching is performed via iterative closest point. Based on our results, we conclude that both structure from motion and shape from shading are viable approaches for 3D ear recognition from video sequences.

AB - Two 3D ear recognition systems using structure from motion (SFM) and shape from shading (SFS) techniques, respectively, are explored. Segmentation of the ear region is performed using interpolation of ridges and ravines identified in each frame in a video sequence. For the SFM system, salient features are tracked across the video sequence and are reconstructed in 3D using a factorization method. Reconstructed points located within the valid ear region are stored as the ear model. The dataset used consists of video sequences for 48 subjects. Each test model is optimally aligned to the database models using a combination of geometric transformations which result in a minimal partial Hausdorff distance. For the SFS system, the ear structure is recovered by using reflectance and illumination properties of the scene. Shape matching is performed via iterative closest point. Based on our results, we conclude that both structure from motion and shape from shading are viable approaches for 3D ear recognition from video sequences.

UR - http://www.scopus.com/inward/record.url?scp=48649098006&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48649098006&partnerID=8YFLogxK

U2 - 10.1109/BTAS.2007.4401938

DO - 10.1109/BTAS.2007.4401938

M3 - Conference contribution

SN - 9781424415977

BT - IEEE Conference on Biometrics: Theory, Applications and Systems, BTAS'07

ER -