DMCA
Electronic Letters on Computer Vision and Image Analysis 5(2):87-97, 2005 Self-supervised adaptation for on-line script text recognition (2004)
BibTeX
@MISC{Prevost04electronicletters,
author = {Lionel Prevost and Loïc Oudot and Universit Pierre and Marie Curie},
title = {Electronic Letters on Computer Vision and Image Analysis 5(2):87-97, 2005 Self-supervised adaptation for on-line script text recognition},
year = {2004}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We have recently developed in our lab a text recognizer for on-line texts written on a touch-terminal. We present in this paper several strategies to adapt this recognizer in a self-supervised way to a given writer and compare them to the supervised adaptation scheme. The baseline system is based on the activationverification cognitive model. We have designed this recognizer to be writer-independent but it may be adapted to be writer-dependent in order to increase the recognition speed and rate. The classification expert can be iteratively modified in order to learn the particularities of a writer. The best self-supervised adaptation strategy is called prototype dynamic management and gets good results, close to those of the supervised methods. The combination of supervised and self-supervised strategies increases accuracy again. Results, presented on a large database of 90 texts (5,400 words) written by 38 different writers are very encouraging with an error rate lower than 10 %. Key Words: handwriting recognition, supervised adaptation, self-supervised adaptation, model-based classifier. 1
Keyphrases
self-supervised adaptation image analysis computer vision electronic letter on-line script text recognition text recognizer self-supervised way model-based classifier handwriting recognition good result baseline system large database supervised method different writer self-supervised strategy classification expert prototype dynamic management adaptation scheme key word paper several strategy on-line text activationverification cognitive model recognition speed error rate self-supervised adaptation strategy
