Utilisation du Web pour la reconnaissance de mots manuscrits hors vocabulaire

Handwriting recognition systems rely on predefined classifiers. Small and static dictionaries are usually exploited to obtain high in-vocabulary (IV) accuracy at the expense of coverage. Thus the recognition of out-of-vocabulary (OOV) words cannot be handled efficiently. To improve OOV recognition while keeping IV dictionaries small, we introduce a multi-step approach that exploits Web resources. After an initial IV-OOV classification, external resources are used to create OOV sequence-adapted dynamic dictionaries. A final CTC-based decoding is performed over the dynamic dictionary to determine the most probable word for the OOV sequence. We validate our approach with experiments conducted on the RIMES dataset. Results show that improvements are obtained compared to standard handwriting recognition. MOTS-CLÉS : reconnaissance d’écriture manuscrite, dictionnaires dynamiques, Wikipédia