95 articles 

inria-00103749, version 1

Personalization of an Online Handwriting Recognition System

Patrick Haluptzok () 1, Michael Revow () 1, Ahmad Abdulkader () 1

Tenth International Workshop on Frontiers in Handwriting Recognition (2006)

Abstract: This paper proposes and compares some approaches for personalizing a handwriting recognizer to a specific user's handwriting style. A typical PocketPC or TabletPC is used by one person exclusively. The handwriting recognizer on such a device can customize its recognition to the specific writing style of the user. This paper presents the results of different personalization approaches for a neural network based classifier, showing how using data specific to the user can dramatically improve recognition accuracy.

  • 1:  Microsoft Corporation
  • Microsoft Corporation
  • Domain : Computer Science/Document and Text Processing
    Computer Science/Computer Vision and Pattern Recognition
  • Keywords : Personalization – Handwriting Recognition – Neural Network
  • Comment : http://www.suvisoft.com
 
  • inria-00103749, version 1
  • http://hal.inria.fr/inria-00103749
  • oai:hal.inria.fr:inria-00103749
  • From: Anne Jaigu <>
  • Submitted on: Thursday, 5 October 2006 11:25:57
  • Updated on: Thursday, 5 October 2006 11:32:14