Deep Learning and Recurrent Connectionist-based Approaches for Arabic Text Recognition in Videos - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Deep Learning and Recurrent Connectionist-based Approaches for Arabic Text Recognition in Videos

Résumé

This paper focuses on recognizing Arabic text embedded in videos. The proposed methods proceed without applying any prior pre-processing operations or character segmentation. Difficulties related to the video or text properties are faced using a learned robust representation of the input text image. This is performed using deep auto-encoders and Convolutional Neural Networks. Features are computed using a multiscale sliding window scheme. A connectionist recurrent approach is then used. It is trained to predict correct transcriptions of an input image from the associated sequence of features. Proposed methods are extensively evaluated on a large database of Arabic TV channels videos and compared to existing solutions.
Fichier non déposé

Dates et versions

hal-01152209 , version 1 (15-05-2015)

Identifiants

  • HAL Id : hal-01152209 , version 1

Citer

Sonia Yousfi, Sid-Ahmed Berrani, Christophe Garcia. Deep Learning and Recurrent Connectionist-based Approaches for Arabic Text Recognition in Videos. 13th International Conference on Document Analysis and Recognition (ICDAR 2015), Aug 2015, Tunis, Tunisia. ⟨hal-01152209⟩
289 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More