From Text Detection in Videos to Person Identification

Johann Poignant 1 Laurent Besacier 1 Georges Quénot 2, * Franck Thollard 2
* Corresponding author
2 MRIM - Modélisation et Recherche d’Information Multimédia [Grenoble]
LIG - Laboratoire d'Informatique de Grenoble, Inria - Institut National de Recherche en Informatique et en Automatique
Abstract : We present in this article a video OCR system that detects and recognizes overlaid texts in video as well as its application to person identification in video documents. We proceed in several steps. First, text detection and temporal tracking are performed. After adaptation of images to a standard OCR system, a final post-processing combines multiple transcriptions of the same text box. The semi-supervised adaptation of this system to a particular video type (video broadcast from a French TV) is proposed and evaluated. The system is efficient as it runs 3 times faster than real time (including the OCR step) on a desktop Linux box. Both text detection and recognition are evaluated individually and through a person recognition task where it is shown that the combination of OCR and audio (speaker) information can greatly improve the performances of a state of the art audio based person identification system.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [19 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00767383
Contributor : Georges Quénot <>
Submitted on : Wednesday, December 19, 2012 - 5:27:16 PM
Last modification on : Tuesday, February 12, 2019 - 1:31:10 AM
Document(s) archivé(s) le : Wednesday, March 20, 2013 - 11:34:16 AM

File

Poignant-Besacier-Quenot-Tholl...
Files produced by the author(s)

Identifiers

Citation

Johann Poignant, Laurent Besacier, Georges Quénot, Franck Thollard. From Text Detection in Videos to Person Identification. ICME 2012 - International Conference on Multimedia and Expo, Jul 2012, Melbourne, VIC, Australia. pp.854-859, ⟨10.1109/ICME.2012.119⟩. ⟨hal-00767383⟩

Share

Metrics

Record views

309

Files downloads

325