Audio-Video Person Authenticate Based on 3D Facial Feature Warping

Girija Chetty, Michael Wagner

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

Abstract

In this paper we propose a novel feature warping technique based on thin-plate-spline (TPS) analysis for 3D audio-video person authentication systems. The TPS warp features model information related to non-rigid variations on speaking faces, such as expression lines, gestures, and wrinkles, enhancing the performance of the system against impostor and spoof attacks. Experiments with multimodal fusion of acoustic and TPS shape features for two different speaking face data corpus, VidTIMIT and AVOZES, allowed equal error rates (EERs) of less than 0.5 % for imposter attacks, less than 1 % for type-1 replay attacks (still photo and pre-recorded audio) and less than 2% for more complex type-2 replay attacks (prerecorded video or fake CG animated video).

Original languageEnglish
Title of host publicationProceedings of the Digital Imaging Computing
Subtitle of host publicationTechniques and Applications, DICTA 2005
EditorsBrian Lovell, Anthony Maeder, Terry Caelli, Sebastian Ourselin
Place of PublicationPiscataway, New Jersey USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages399-406
Number of pages8
Volume2005
ISBN (Print)0769524672, 9780769524672
DOIs
Publication statusPublished - 2005
EventDigital Imaging Computing: Techniques and Applications, DICTA 2005 - Cairns, Australia
Duration: 6 Dec 20058 Dec 2005

Conference

ConferenceDigital Imaging Computing: Techniques and Applications, DICTA 2005
CountryAustralia
CityCairns
Period6/12/058/12/05

Fingerprint

Splines
Authentication
Fusion reactions
Acoustics
Experiments

Cite this

Chetty, G., & Wagner, M. (2005). Audio-Video Person Authenticate Based on 3D Facial Feature Warping. In B. Lovell, A. Maeder, T. Caelli, & S. Ourselin (Eds.), Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005 (Vol. 2005, pp. 399-406). [1578156] Piscataway, New Jersey USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/DICTA.2005.13
Chetty, Girija ; Wagner, Michael. / Audio-Video Person Authenticate Based on 3D Facial Feature Warping. Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. editor / Brian Lovell ; Anthony Maeder ; Terry Caelli ; Sebastian Ourselin. Vol. 2005 Piscataway, New Jersey USA : IEEE, Institute of Electrical and Electronics Engineers, 2005. pp. 399-406
@inproceedings{37b807ac18ed41da9c14946eef5550c7,
title = "Audio-Video Person Authenticate Based on 3D Facial Feature Warping",
abstract = "In this paper we propose a novel feature warping technique based on thin-plate-spline (TPS) analysis for 3D audio-video person authentication systems. The TPS warp features model information related to non-rigid variations on speaking faces, such as expression lines, gestures, and wrinkles, enhancing the performance of the system against impostor and spoof attacks. Experiments with multimodal fusion of acoustic and TPS shape features for two different speaking face data corpus, VidTIMIT and AVOZES, allowed equal error rates (EERs) of less than 0.5 {\%} for imposter attacks, less than 1 {\%} for type-1 replay attacks (still photo and pre-recorded audio) and less than 2{\%} for more complex type-2 replay attacks (prerecorded video or fake CG animated video).",
keywords = "face recognition",
author = "Girija Chetty and Michael Wagner",
year = "2005",
doi = "10.1109/DICTA.2005.13",
language = "English",
isbn = "0769524672",
volume = "2005",
pages = "399--406",
editor = "Brian Lovell and Anthony Maeder and Terry Caelli and Sebastian Ourselin",
booktitle = "Proceedings of the Digital Imaging Computing",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Chetty, G & Wagner, M 2005, Audio-Video Person Authenticate Based on 3D Facial Feature Warping. in B Lovell, A Maeder, T Caelli & S Ourselin (eds), Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. vol. 2005, 1578156, IEEE, Institute of Electrical and Electronics Engineers, Piscataway, New Jersey USA, pp. 399-406, Digital Imaging Computing: Techniques and Applications, DICTA 2005, Cairns, Australia, 6/12/05. https://doi.org/10.1109/DICTA.2005.13

Audio-Video Person Authenticate Based on 3D Facial Feature Warping. / Chetty, Girija; Wagner, Michael.

Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. ed. / Brian Lovell; Anthony Maeder; Terry Caelli; Sebastian Ourselin. Vol. 2005 Piscataway, New Jersey USA : IEEE, Institute of Electrical and Electronics Engineers, 2005. p. 399-406 1578156.

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

TY - GEN

T1 - Audio-Video Person Authenticate Based on 3D Facial Feature Warping

AU - Chetty, Girija

AU - Wagner, Michael

PY - 2005

Y1 - 2005

N2 - In this paper we propose a novel feature warping technique based on thin-plate-spline (TPS) analysis for 3D audio-video person authentication systems. The TPS warp features model information related to non-rigid variations on speaking faces, such as expression lines, gestures, and wrinkles, enhancing the performance of the system against impostor and spoof attacks. Experiments with multimodal fusion of acoustic and TPS shape features for two different speaking face data corpus, VidTIMIT and AVOZES, allowed equal error rates (EERs) of less than 0.5 % for imposter attacks, less than 1 % for type-1 replay attacks (still photo and pre-recorded audio) and less than 2% for more complex type-2 replay attacks (prerecorded video or fake CG animated video).

AB - In this paper we propose a novel feature warping technique based on thin-plate-spline (TPS) analysis for 3D audio-video person authentication systems. The TPS warp features model information related to non-rigid variations on speaking faces, such as expression lines, gestures, and wrinkles, enhancing the performance of the system against impostor and spoof attacks. Experiments with multimodal fusion of acoustic and TPS shape features for two different speaking face data corpus, VidTIMIT and AVOZES, allowed equal error rates (EERs) of less than 0.5 % for imposter attacks, less than 1 % for type-1 replay attacks (still photo and pre-recorded audio) and less than 2% for more complex type-2 replay attacks (prerecorded video or fake CG animated video).

KW - face recognition

UR - http://www.scopus.com/inward/record.url?scp=33846990425&partnerID=8YFLogxK

UR - http://dicta2005.aprs.org.au/

U2 - 10.1109/DICTA.2005.13

DO - 10.1109/DICTA.2005.13

M3 - Conference contribution

SN - 0769524672

SN - 9780769524672

VL - 2005

SP - 399

EP - 406

BT - Proceedings of the Digital Imaging Computing

A2 - Lovell, Brian

A2 - Maeder, Anthony

A2 - Caelli, Terry

A2 - Ourselin, Sebastian

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway, New Jersey USA

ER -

Chetty G, Wagner M. Audio-Video Person Authenticate Based on 3D Facial Feature Warping. In Lovell B, Maeder A, Caelli T, Ourselin S, editors, Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. Vol. 2005. Piscataway, New Jersey USA: IEEE, Institute of Electrical and Electronics Engineers. 2005. p. 399-406. 1578156 https://doi.org/10.1109/DICTA.2005.13