Online Sketch Recognition with Incremental Fuzzy Models

Abstract : In this paper, we present a novel method for recognition of handwritten sketches. Unlike previous approaches, we focus on online retrieval and ability to build our model incrementally, thus we do not need to know all the data in advance and we can achieve very good recognition results after as few as 15 samples. The method is composed of two main parts: feature representation and learning and recognition. In feature representation part, we utilize SIFT-like feature descriptors in combination with soft response Bag-of-Words techniques. Descriptors are extracted locally using our novel sketch-specific sampling strategy and for support regions we follow patch-based approach. For learning and recognition, we use a novel technique based on fuzzy-neural networks, which has shown good performance in incremental learning. The experiments on state-of-the-art benchmarks have shown promising results.
Type de document :
Communication dans un congrès
Céline Rémi; Lionel Prévost; Eric Anquetil. 17th Biennial Conference of the International Graphonomics Society, Jun 2015, Pointe-à-Pitre, Guadeloupe. 2015, Drawing, Handwriting Processing Analysis: New Advances and Challenges
Liste complète des métadonnées

https://hal.univ-antilles.fr/hal-01166497
Contributeur : Philippe Simon <>
Soumis le : lundi 22 juin 2015 - 20:44:55
Dernière modification le : mardi 23 juin 2015 - 14:58:53
Document(s) archivé(s) le : mardi 15 septembre 2015 - 20:55:47

Fichier

IGS_2015_submission_37.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01166497, version 1

Collections

Citation

Lukas Tencer, Marta Režnáková, Mohamed Cheriet. Online Sketch Recognition with Incremental Fuzzy Models. Céline Rémi; Lionel Prévost; Eric Anquetil. 17th Biennial Conference of the International Graphonomics Society, Jun 2015, Pointe-à-Pitre, Guadeloupe. 2015, Drawing, Handwriting Processing Analysis: New Advances and Challenges. 〈hal-01166497〉

Partager

Métriques

Consultations de la notice

258

Téléchargements de fichiers

121