Fusion of speech, faces and text for person identification in TV broadcast

Hervé Bredin; Johann Poignant; Makarand Tapaswi; Guillaume Fortier; Viet Bac Le; Thibault Napoleon; Hua Gao; Claude Barras; Sophie Rosset; Laurent Besacier; Jakob Verbeek; Georges Quénot; Frédéric Jurie; Hazim Kemal Ekenel

doi:10.1007/978-3-642-33885-4_39

Fusion of speech, faces and text for person identification in TV broadcast

Hervé Bredin^*, Johann Poignant, Makarand Tapaswi, Guillaume Fortier, Viet Bac Le, Thibault Napoleon, Hua Gao, Claude Barras, Sophie Rosset, Laurent Besacier, Jakob Verbeek, Georges Quénot, Frédéric Jurie, Hazim Kemal Ekenel

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

9 Citations (Scopus)

Abstract

The Repere challenge is a project aiming at the evaluation of systems for supervised and unsupervised multimodal recognition of people in TV broadcast. In this paper, we describe, evaluate and discuss QCompere consortium submissions to the 2012 Repere evaluation campaign dry-run. Speaker identification (and face recognition) can be greatly improved when combined with name detection through video optical character recognition. Moreover, we show that unsupervised multimodal person recognition systems can achieve performance nearly as good as supervised monomodal ones (with several hundreds of identity models).

Original language	English
Title of host publication	Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings
Publisher	Springer Verlag
Pages	385-394
Number of pages	10
Edition	PART 3
ISBN (Print)	9783642338847
DOIs	https://doi.org/10.1007/978-3-642-33885-4_39
Publication status	Published - 2012
Externally published	Yes
Event	Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings - Florence, Italy Duration: 7 Oct 2012 → 13 Oct 2012

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 3
Volume	7585 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings
Country/Territory	Italy
City	Florence
Period	7/10/12 → 13/10/12

Access to Document

10.1007/978-3-642-33885-4_39

Cite this

Bredin, H., Poignant, J., Tapaswi, M., Fortier, G., Le, V. B., Napoleon, T., Gao, H., Barras, C., Rosset, S., Besacier, L., Verbeek, J., Quénot, G., Jurie, F., & Ekenel, H. K. (2012). Fusion of speech, faces and text for person identification in TV broadcast. In Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings (PART 3 ed., pp. 385-394). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7585 LNCS, No. PART 3). Springer Verlag. https://doi.org/10.1007/978-3-642-33885-4_39

Bredin, Hervé ; Poignant, Johann ; Tapaswi, Makarand et al. / Fusion of speech, faces and text for person identification in TV broadcast. Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings. PART 3. ed. Springer Verlag, 2012. pp. 385-394 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3).

@inproceedings{d3050cbe97ec47c98e4af2327035ae6c,

title = "Fusion of speech, faces and text for person identification in TV broadcast",

abstract = "The Repere challenge is a project aiming at the evaluation of systems for supervised and unsupervised multimodal recognition of people in TV broadcast. In this paper, we describe, evaluate and discuss QCompere consortium submissions to the 2012 Repere evaluation campaign dry-run. Speaker identification (and face recognition) can be greatly improved when combined with name detection through video optical character recognition. Moreover, we show that unsupervised multimodal person recognition systems can achieve performance nearly as good as supervised monomodal ones (with several hundreds of identity models).",

author = "Herv{\'e} Bredin and Johann Poignant and Makarand Tapaswi and Guillaume Fortier and Le, {Viet Bac} and Thibault Napoleon and Hua Gao and Claude Barras and Sophie Rosset and Laurent Besacier and Jakob Verbeek and Georges Qu{\'e}not and Fr{\'e}d{\'e}ric Jurie and Ekenel, {Hazim Kemal}",

year = "2012",

doi = "10.1007/978-3-642-33885-4_39",

language = "English",

isbn = "9783642338847",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

number = "PART 3",

pages = "385--394",

booktitle = "Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings",

address = "Germany",

edition = "PART 3",

note = "Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings ; Conference date: 07-10-2012 Through 13-10-2012",

}

Bredin, H, Poignant, J, Tapaswi, M, Fortier, G, Le, VB, Napoleon, T, Gao, H, Barras, C, Rosset, S, Besacier, L, Verbeek, J, Quénot, G, Jurie, F & Ekenel, HK 2012, Fusion of speech, faces and text for person identification in TV broadcast. in Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings. PART 3 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 3, vol. 7585 LNCS, Springer Verlag, pp. 385-394, Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings, Florence, Italy, 7/10/12. https://doi.org/10.1007/978-3-642-33885-4_39

Fusion of speech, faces and text for person identification in TV broadcast. / Bredin, Hervé; Poignant, Johann; Tapaswi, Makarand et al.
Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings. PART 3. ed. Springer Verlag, 2012. p. 385-394 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7585 LNCS, No. PART 3).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Fusion of speech, faces and text for person identification in TV broadcast

AU - Bredin, Hervé

AU - Poignant, Johann

AU - Tapaswi, Makarand

AU - Fortier, Guillaume

AU - Le, Viet Bac

AU - Napoleon, Thibault

AU - Gao, Hua

AU - Barras, Claude

AU - Rosset, Sophie

AU - Besacier, Laurent

AU - Verbeek, Jakob

AU - Quénot, Georges

AU - Jurie, Frédéric

AU - Ekenel, Hazim Kemal

PY - 2012

Y1 - 2012

N2 - The Repere challenge is a project aiming at the evaluation of systems for supervised and unsupervised multimodal recognition of people in TV broadcast. In this paper, we describe, evaluate and discuss QCompere consortium submissions to the 2012 Repere evaluation campaign dry-run. Speaker identification (and face recognition) can be greatly improved when combined with name detection through video optical character recognition. Moreover, we show that unsupervised multimodal person recognition systems can achieve performance nearly as good as supervised monomodal ones (with several hundreds of identity models).

AB - The Repere challenge is a project aiming at the evaluation of systems for supervised and unsupervised multimodal recognition of people in TV broadcast. In this paper, we describe, evaluate and discuss QCompere consortium submissions to the 2012 Repere evaluation campaign dry-run. Speaker identification (and face recognition) can be greatly improved when combined with name detection through video optical character recognition. Moreover, we show that unsupervised multimodal person recognition systems can achieve performance nearly as good as supervised monomodal ones (with several hundreds of identity models).

UR - http://www.scopus.com/inward/record.url?scp=84867719972&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-33885-4_39

DO - 10.1007/978-3-642-33885-4_39

M3 - Conference contribution

AN - SCOPUS:84867719972

SN - 9783642338847

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 385

EP - 394

BT - Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings

PB - Springer Verlag

T2 - Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings

Y2 - 7 October 2012 through 13 October 2012

ER -

Bredin H, Poignant J, Tapaswi M, Fortier G, Le VB, Napoleon T et al. Fusion of speech, faces and text for person identification in TV broadcast. In Computer Vision, ECCV 2012 - Workshops and Demonstrations, Proceedings. PART 3 ed. Springer Verlag. 2012. p. 385-394. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3). doi: 10.1007/978-3-642-33885-4_39

Fusion of speech, faces and text for person identification in TV broadcast

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this