Fully automatic face recognition from 3D videos

Munawar Hayat, Mohammed Bennamoun, Amar El-Sallam

Research output: A Conference proceeding or a Chapter in BookConference contribution

5 Citations (Scopus)
2 Downloads (Pure)

Abstract

Almost all of the existing research on 3D face recognition is based on static 3D images. 3D videos are believed to provide more information in terms of both the shape as well as the dynamics of an individual's face. This paper presents a system which exploits the spatiotemporal information in 3D videos for the task of face recognition. An algorithm for automatic normalization of raw 3D videos is also given. After the detection of the nose tip, all meshes of the 3D video are cropped and uniformly sampled to form range videos. Spatiotemporal Local Binary Pattern (LBP) descriptors are used for feature extraction from the range videos. For classification, a linear multiclass Support Vector Machine (SVM) is used. The system is trained on videos of a person with different facial expressions and tested on a video with new facial expression. Experimental results on the largest currently available 3D video database, BU 4DFE, show a high recognition rate of 92.68%.
Original languageEnglish
Title of host publication2012 21st International conference on Pattern recognition (ICPR 2012)
EditorsJan-Olof Eklundh, Yuichi Ohta, Steven Tanimoto
Place of PublicationTskuba Japan
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1415-1418
Number of pages4
ISBN (Electronic)9784990644109
ISBN (Print)9781467322164
Publication statusPublished - 2012
Externally publishedYes
Event21st International Conference on Pattern Recognition (ICPR 2012) - Tsukuba, Tsukuba, Japan
Duration: 11 Nov 201215 Nov 2012

Conference

Conference21st International Conference on Pattern Recognition (ICPR 2012)
Abbreviated titleICPR 2012
CountryJapan
CityTsukuba
Period11/11/1215/11/12

Fingerprint

Face recognition
Support vector machines
Feature extraction

Cite this

Hayat, M., Bennamoun, M., & El-Sallam, A. (2012). Fully automatic face recognition from 3D videos. In J-O. Eklundh, Y. Ohta, & S. Tanimoto (Eds.), 2012 21st International conference on Pattern recognition (ICPR 2012) (pp. 1415-1418). Tskuba Japan: IEEE, Institute of Electrical and Electronics Engineers.
Hayat, Munawar ; Bennamoun, Mohammed ; El-Sallam, Amar. / Fully automatic face recognition from 3D videos. 2012 21st International conference on Pattern recognition (ICPR 2012). editor / Jan-Olof Eklundh ; Yuichi Ohta ; Steven Tanimoto. Tskuba Japan : IEEE, Institute of Electrical and Electronics Engineers, 2012. pp. 1415-1418
@inproceedings{141109295c0a491780e78bfe92008088,
title = "Fully automatic face recognition from 3D videos",
abstract = "Almost all of the existing research on 3D face recognition is based on static 3D images. 3D videos are believed to provide more information in terms of both the shape as well as the dynamics of an individual's face. This paper presents a system which exploits the spatiotemporal information in 3D videos for the task of face recognition. An algorithm for automatic normalization of raw 3D videos is also given. After the detection of the nose tip, all meshes of the 3D video are cropped and uniformly sampled to form range videos. Spatiotemporal Local Binary Pattern (LBP) descriptors are used for feature extraction from the range videos. For classification, a linear multiclass Support Vector Machine (SVM) is used. The system is trained on videos of a person with different facial expressions and tested on a video with new facial expression. Experimental results on the largest currently available 3D video database, BU 4DFE, show a high recognition rate of 92.68{\%}.",
keywords = "Face-recognition, Pattern-recognition, computer-vision",
author = "Munawar Hayat and Mohammed Bennamoun and Amar El-Sallam",
year = "2012",
language = "English",
isbn = "9781467322164",
pages = "1415--1418",
editor = "Jan-Olof Eklundh and Yuichi Ohta and Steven Tanimoto",
booktitle = "2012 21st International conference on Pattern recognition (ICPR 2012)",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Hayat, M, Bennamoun, M & El-Sallam, A 2012, Fully automatic face recognition from 3D videos. in J-O Eklundh, Y Ohta & S Tanimoto (eds), 2012 21st International conference on Pattern recognition (ICPR 2012). IEEE, Institute of Electrical and Electronics Engineers, Tskuba Japan, pp. 1415-1418, 21st International Conference on Pattern Recognition (ICPR 2012), Tsukuba, Japan, 11/11/12.

Fully automatic face recognition from 3D videos. / Hayat, Munawar; Bennamoun, Mohammed; El-Sallam, Amar.

2012 21st International conference on Pattern recognition (ICPR 2012). ed. / Jan-Olof Eklundh; Yuichi Ohta; Steven Tanimoto. Tskuba Japan : IEEE, Institute of Electrical and Electronics Engineers, 2012. p. 1415-1418.

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Fully automatic face recognition from 3D videos

AU - Hayat, Munawar

AU - Bennamoun, Mohammed

AU - El-Sallam, Amar

PY - 2012

Y1 - 2012

N2 - Almost all of the existing research on 3D face recognition is based on static 3D images. 3D videos are believed to provide more information in terms of both the shape as well as the dynamics of an individual's face. This paper presents a system which exploits the spatiotemporal information in 3D videos for the task of face recognition. An algorithm for automatic normalization of raw 3D videos is also given. After the detection of the nose tip, all meshes of the 3D video are cropped and uniformly sampled to form range videos. Spatiotemporal Local Binary Pattern (LBP) descriptors are used for feature extraction from the range videos. For classification, a linear multiclass Support Vector Machine (SVM) is used. The system is trained on videos of a person with different facial expressions and tested on a video with new facial expression. Experimental results on the largest currently available 3D video database, BU 4DFE, show a high recognition rate of 92.68%.

AB - Almost all of the existing research on 3D face recognition is based on static 3D images. 3D videos are believed to provide more information in terms of both the shape as well as the dynamics of an individual's face. This paper presents a system which exploits the spatiotemporal information in 3D videos for the task of face recognition. An algorithm for automatic normalization of raw 3D videos is also given. After the detection of the nose tip, all meshes of the 3D video are cropped and uniformly sampled to form range videos. Spatiotemporal Local Binary Pattern (LBP) descriptors are used for feature extraction from the range videos. For classification, a linear multiclass Support Vector Machine (SVM) is used. The system is trained on videos of a person with different facial expressions and tested on a video with new facial expression. Experimental results on the largest currently available 3D video database, BU 4DFE, show a high recognition rate of 92.68%.

KW - Face-recognition

KW - Pattern-recognition

KW - computer-vision

M3 - Conference contribution

SN - 9781467322164

SP - 1415

EP - 1418

BT - 2012 21st International conference on Pattern recognition (ICPR 2012)

A2 - Eklundh, Jan-Olof

A2 - Ohta, Yuichi

A2 - Tanimoto, Steven

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Tskuba Japan

ER -

Hayat M, Bennamoun M, El-Sallam A. Fully automatic face recognition from 3D videos. In Eklundh J-O, Ohta Y, Tanimoto S, editors, 2012 21st International conference on Pattern recognition (ICPR 2012). Tskuba Japan: IEEE, Institute of Electrical and Electronics Engineers. 2012. p. 1415-1418