Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication

Girija Chetty; Michael Wagner

doi:10.1109/ISSPA.2005.1580197

Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication

Girija Chetty, Michael Wagner

Information Systems

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

9 Citations (Scopus)

49 Downloads (Pure)

Abstract

In this paper we propose a feature level fusion approach for checking liveness in face-voice person authentication. Liveness verification experiments conducted on two audiovisual databases, VidTIMIT and UCBN, show that feature-level fusion is indeed a powerful technique for checking liveness in systems that are vulnerable to replay attacks, as it preserves synchronisation between closely coupled modalities, such as voice and face, through various stages of authentication. An improvement in error rate of the order of 25-40% is achieved for replay attack experiments by using feature level fusion of acoustic and visual feature vectors from lip region as compared to classical late fusion approach.

Original language	English
Title of host publication	Proceedings of the Eighth International Symposium on Signal Processing and Applications
Place of Publication	Piscataway, New Jersey, USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	66-69
Number of pages	4
ISBN (Print)	0-7803-9244-2
DOIs	https://doi.org/10.1109/ISSPA.2005.1580197
Publication status	Published - 2005
Event	ISSPA-2005 - Sydney, Australia Duration: 28 Aug 2005 → 31 Aug 2005

Conference

Conference	ISSPA-2005
Country/Territory	Australia
City	Sydney
Period	28/08/05 → 31/08/05

Access to Document

10.1109/ISSPA.2005.1580197

PaperFinal published version, 341 KB

Cite this

@inproceedings{8336f93ced604b388f80ae94da03a5c2,

title = "Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication",

abstract = "In this paper we propose a feature level fusion approach for checking liveness in face-voice person authentication. Liveness verification experiments conducted on two audiovisual databases, VidTIMIT and UCBN, show that feature-level fusion is indeed a powerful technique for checking liveness in systems that are vulnerable to replay attacks, as it preserves synchronisation between closely coupled modalities, such as voice and face, through various stages of authentication. An improvement in error rate of the order of 25-40% is achieved for replay attack experiments by using feature level fusion of acoustic and visual feature vectors from lip region as compared to classical late fusion approach.",

author = "Girija Chetty and Michael Wagner",

year = "2005",

doi = "10.1109/ISSPA.2005.1580197",

language = "English",

isbn = "0-7803-9244-2",

pages = "66--69",

booktitle = "Proceedings of the Eighth International Symposium on Signal Processing and Applications",

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

address = "United States",

note = "ISSPA-2005 ; Conference date: 28-08-2005 Through 31-08-2005",

}

Chetty, G & Wagner, M 2005, Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication. in Proceedings of the Eighth International Symposium on Signal Processing and Applications. IEEE, Institute of Electrical and Electronics Engineers, Piscataway, New Jersey, USA, pp. 66-69, ISSPA-2005, Sydney, Australia, 28/08/05. https://doi.org/10.1109/ISSPA.2005.1580197

Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication. / Chetty, Girija; Wagner, Michael.
Proceedings of the Eighth International Symposium on Signal Processing and Applications. Piscataway, New Jersey, USA: IEEE, Institute of Electrical and Electronics Engineers, 2005. p. 66-69.

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

TY - GEN

T1 - Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication

AU - Chetty, Girija

AU - Wagner, Michael

PY - 2005

Y1 - 2005

N2 - In this paper we propose a feature level fusion approach for checking liveness in face-voice person authentication. Liveness verification experiments conducted on two audiovisual databases, VidTIMIT and UCBN, show that feature-level fusion is indeed a powerful technique for checking liveness in systems that are vulnerable to replay attacks, as it preserves synchronisation between closely coupled modalities, such as voice and face, through various stages of authentication. An improvement in error rate of the order of 25-40% is achieved for replay attack experiments by using feature level fusion of acoustic and visual feature vectors from lip region as compared to classical late fusion approach.

AB - In this paper we propose a feature level fusion approach for checking liveness in face-voice person authentication. Liveness verification experiments conducted on two audiovisual databases, VidTIMIT and UCBN, show that feature-level fusion is indeed a powerful technique for checking liveness in systems that are vulnerable to replay attacks, as it preserves synchronisation between closely coupled modalities, such as voice and face, through various stages of authentication. An improvement in error rate of the order of 25-40% is achieved for replay attack experiments by using feature level fusion of acoustic and visual feature vectors from lip region as compared to classical late fusion approach.

U2 - 10.1109/ISSPA.2005.1580197

DO - 10.1109/ISSPA.2005.1580197

M3 - Conference contribution

SN - 0-7803-9244-2

SP - 66

EP - 69

BT - Proceedings of the Eighth International Symposium on Signal Processing and Applications

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway, New Jersey, USA

T2 - ISSPA-2005

Y2 - 28 August 2005 through 31 August 2005

ER -

Investigating Feature Level Fusion for Checking Liveness in Face-Voice Authentication

Abstract

Conference

Access to Document

Fingerprint

Cite this