inria-00103797, version 2

Unsupervised Estimation of Writing Style Models for Improved Unconstrained Off-line Handwriting Recognition

Gernot A. Fink () ¹, Thomas Plötz () ²

Tenth International Workshop on Frontiers in Handwriting Recognition (2006)

• 1:  Robotics Research Institute - Intelligent Systems Group
• 
  University of Dortmund Germany
• 2:  Faculty of Technology - Applied Computer Science Group
• 
  Bielefeld University Germany

Available versions :  v1 (2006-10-05) v2 (2006-11-21)

Bibliographic reference

• Type of document: Peer-reviewed conferences/proceedings
• Domain:

  Computer Science/Computer Vision and Pattern Recognition

  Computer Science/Document and Text Processing

• Title: Unsupervised Estimation of Writing Style Models for Improved Unconstrained Off-line Handwriting Recognition
• Abstract: The performance of writer-independent unconstrained handwriting recognition is severely affected by variations in writing style. In a segmentation-free approach based on Hidden-Markov models we, therefore, use multiple recognition models specialized to specific writing styles in order to improve recognition performance. As the explicit definition of writing styles is not obvious we propose an unsupervised clustering procedure that estimates Gaussian mixture models for writing styles in a completely datadriven manner and thus implicitly establishes classes of writing styles. On a challenging writer-independent unconstrained handwriting recognition task our two stage recognition approach – first performing a writing style classification and then using a style-specific writing model for decoding – achieves superior performance compared to a single style-independent baseline system.
• ACM Classification:

  I.: Computing Methodologies/I.5: PATTERN RECOGNITION

  I.: Computing Methodologies/I.7: DOCUMENT AND TEXT PROCESSING

• Full text language: English
• Publication date: 2006-10-23
• Audience: not specified
• Conference title: Tenth International Workshop on Frontiers in Handwriting Recognition
• Conference city: La Baule (France)
• Conference date: 2006-10-23
• Organizer: Université de Rennes 1
• Scientific editor(s): Guy Lorette
• Commercial editor: Suvisoft
• Keywords: Unconstrained handwriting – segmentation-fre recognition – writing-style model
• Comment: http://www.suvisoft.com
• Contract, financing: Université de Rennes 1

Attached file list to this document: 

• inria-00103797, version 2
• http://hal.inria.fr/inria-00103797
• oai:hal.inria.fr:inria-00103797
• From: Anne Jaigu <>
• Submitted on: Tuesday, 21 November 2006 09:11:14
• Updated on: Tuesday, 21 November 2006 10:18:15