Matching Tensors for Pose Invariant Automatic 3D Face Recognition

Ajmal Mian; Mohammed Bennamoun; Robyn Owens

doi:10.1109/CVPR.2005.490

Matching Tensors for Pose Invariant Automatic 3D Face Recognition

Ajmal Mian, Mohammed Bennamoun, Robyn Owens

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

11 Citations (Scopus)

Abstract

The face is an easily collectible and non-intrusive biometric used for the authentication and identification of individuals. 2D face recognition techniques are sensitive to changes in illumination, makeup and pose. We present a fully automatic 3D face recognition algorithm that overcomes these limitations. During the enrollment, 3D faces in the gallery are represented by third order tensors which are indexed by a 4D hash table. During online recognition, tensors are computed for a probe and are used to cast votes to the tensors in the gallery using the hash table. Gallery faces are ranked according to their votes and a similarity measure based on a linear correlation coefficient and registration error is calculated only for the high ranked faces. The face with the highest similarity is declared as the recognized face. Experiments were performed on a database of 277 subjects and a rank one recognition rate of 86.4% was achieved. Our results also show that our algorithm’s execution time is insensitive to the gallery size.

Original language	English
Title of host publication	Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society
Editors	Cordelia Schmid, Stefano Soatto, Carlo Tomasi
Place of Publication	Los Alamitos, California, USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	120
ISBN (Print)	0636919
DOIs	https://doi.org/10.1109/CVPR.2005.490
Publication status	Published - Jun 2005
Event	Matching Tensors for Pose Invariant Automatic 3D Face Recognition - San Diego, California, USA Duration: 1 Jan 2005 → …

Conference

Conference	Matching Tensors for Pose Invariant Automatic 3D Face Recognition
Period	1/01/05 → …

Access to Document

10.1109/CVPR.2005.490

An Automated 3D Model-based Object Recognition System
Bennamoun, M.
ARC Discovery Projects
1/01/03 → 31/12/05
Project: Research

Cite this

Mian, A., Bennamoun, M., & Owens, R. (2005). Matching Tensors for Pose Invariant Automatic 3D Face Recognition. In C. Schmid, S. Soatto, & C. Tomasi (Eds.), Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society (pp. 120). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CVPR.2005.490

Mian, Ajmal ; Bennamoun, Mohammed ; Owens, Robyn. / Matching Tensors for Pose Invariant Automatic 3D Face Recognition. Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society . editor / Cordelia Schmid ; Stefano Soatto ; Carlo Tomasi. Los Alamitos, California, USA : IEEE, Institute of Electrical and Electronics Engineers, 2005. pp. 120

@inproceedings{b3a859ccd231440c86dc1e2dfad1a324,

title = "Matching Tensors for Pose Invariant Automatic 3D Face Recognition",

abstract = "The face is an easily collectible and non-intrusive biometric used for the authentication and identification of individuals. 2D face recognition techniques are sensitive to changes in illumination, makeup and pose. We present a fully automatic 3D face recognition algorithm that overcomes these limitations. During the enrollment, 3D faces in the gallery are represented by third order tensors which are indexed by a 4D hash table. During online recognition, tensors are computed for a probe and are used to cast votes to the tensors in the gallery using the hash table. Gallery faces are ranked according to their votes and a similarity measure based on a linear correlation coefficient and registration error is calculated only for the high ranked faces. The face with the highest similarity is declared as the recognized face. Experiments were performed on a database of 277 subjects and a rank one recognition rate of 86.4% was achieved. Our results also show that our algorithm{\textquoteright}s execution time is insensitive to the gallery size.",

author = "Ajmal Mian and Mohammed Bennamoun and Robyn Owens",

year = "2005",

month = jun,

doi = "10.1109/CVPR.2005.490",

language = "English",

isbn = "0636919",

pages = "120",

editor = "Cordelia Schmid and Stefano Soatto and Carlo Tomasi",

booktitle = "Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States",

note = "Matching Tensors for Pose Invariant Automatic 3D Face Recognition ; Conference date: 01-01-2005",

}

Mian, A , Bennamoun, M & Owens, R 2005, Matching Tensors for Pose Invariant Automatic 3D Face Recognition. in C Schmid, S Soatto & C Tomasi (eds), Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society . IEEE, Institute of Electrical and Electronics Engineers, Los Alamitos, California, USA, pp. 120, Matching Tensors for Pose Invariant Automatic 3D Face Recognition, 1/01/05. https://doi.org/10.1109/CVPR.2005.490

Matching Tensors for Pose Invariant Automatic 3D Face Recognition. / Mian, Ajmal ; Bennamoun, Mohammed ; Owens, Robyn.
Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society . ed. / Cordelia Schmid; Stefano Soatto; Carlo Tomasi. Los Alamitos, California, USA: IEEE, Institute of Electrical and Electronics Engineers, 2005. p. 120.

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

TY - GEN

T1 - Matching Tensors for Pose Invariant Automatic 3D Face Recognition

AU - Mian, Ajmal

AU - Bennamoun, Mohammed

AU - Owens, Robyn

PY - 2005/6

Y1 - 2005/6

N2 - The face is an easily collectible and non-intrusive biometric used for the authentication and identification of individuals. 2D face recognition techniques are sensitive to changes in illumination, makeup and pose. We present a fully automatic 3D face recognition algorithm that overcomes these limitations. During the enrollment, 3D faces in the gallery are represented by third order tensors which are indexed by a 4D hash table. During online recognition, tensors are computed for a probe and are used to cast votes to the tensors in the gallery using the hash table. Gallery faces are ranked according to their votes and a similarity measure based on a linear correlation coefficient and registration error is calculated only for the high ranked faces. The face with the highest similarity is declared as the recognized face. Experiments were performed on a database of 277 subjects and a rank one recognition rate of 86.4% was achieved. Our results also show that our algorithm’s execution time is insensitive to the gallery size.

AB - The face is an easily collectible and non-intrusive biometric used for the authentication and identification of individuals. 2D face recognition techniques are sensitive to changes in illumination, makeup and pose. We present a fully automatic 3D face recognition algorithm that overcomes these limitations. During the enrollment, 3D faces in the gallery are represented by third order tensors which are indexed by a 4D hash table. During online recognition, tensors are computed for a probe and are used to cast votes to the tensors in the gallery using the hash table. Gallery faces are ranked according to their votes and a similarity measure based on a linear correlation coefficient and registration error is calculated only for the high ranked faces. The face with the highest similarity is declared as the recognized face. Experiments were performed on a database of 277 subjects and a rank one recognition rate of 86.4% was achieved. Our results also show that our algorithm’s execution time is insensitive to the gallery size.

U2 - 10.1109/CVPR.2005.490

DO - 10.1109/CVPR.2005.490

M3 - Conference paper

SN - 0636919

SP - 120

BT - Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society

A2 - Schmid, Cordelia

A2 - Soatto, Stefano

A2 - Tomasi, Carlo

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Los Alamitos, California, USA

T2 - Matching Tensors for Pose Invariant Automatic 3D Face Recognition

Y2 - 1 January 2005

ER -

Mian A , Bennamoun M , Owens R. Matching Tensors for Pose Invariant Automatic 3D Face Recognition. In Schmid C, Soatto S, Tomasi C, editors, Conference on Computer Vision and Pattern Recognition - Workshops, 2005. CVPR Workshops. IEEE Computer Society . Los Alamitos, California, USA: IEEE, Institute of Electrical and Electronics Engineers. 2005. p. 120 doi: 10.1109/CVPR.2005.490

Matching Tensors for Pose Invariant Automatic 3D Face Recognition

Abstract

Conference

Access to Document

Fingerprint

Projects

An Automated 3D Model-based Object Recognition System

Cite this