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).
Cite
Text
Bredin et al. "Fusion of Speech, Faces and Text for Person Identification in TV Broadcast." European Conference on Computer Vision, 2012. doi:10.1007/978-3-642-33885-4_39Markdown
[Bredin et al. "Fusion of Speech, Faces and Text for Person Identification in TV Broadcast." European Conference on Computer Vision, 2012.](https://mlanthology.org/eccv/2012/bredin2012eccv-fusion/) doi:10.1007/978-3-642-33885-4_39BibTeX
@inproceedings{bredin2012eccv-fusion,
title = {{Fusion of Speech, Faces and Text for Person Identification in TV Broadcast}},
author = {Bredin, Hervé and Poignant, Johann and Tapaswi, Makarand and Fortier, Guillaume and Le, Viet Bac and Napoléon, Thibault and Gao, Hua and Barras, Claude and Rosset, Sophie and Besacier, Laurent and Verbeek, Jakob and Quénot, Georges and Jurie, Frédéric and Ekenel, Hazim Kemal},
booktitle = {European Conference on Computer Vision},
year = {2012},
pages = {385-394},
doi = {10.1007/978-3-642-33885-4_39},
url = {https://mlanthology.org/eccv/2012/bredin2012eccv-fusion/}
}