Training of On-line Handwriting Text Recognizers with Synthetic Text Generated Using the Kinematic Theory of Rapid Human Movements

被引：8

作者：

Martin-Albo, Daniel ^{[1
]}

Plamondon, Rejean ^{[2
]}

Vidal, Enrique ^{[1
]}

机构：

[1] Univ Politecn Valencia, PRHLT Res Ctr, E-46022 Valencia, Spain

[2] Ecole Polytech, Lab Scribens, Montreal, PQ, Canada

来源：

2014 14TH INTERNATIONAL CONFERENCE ON FRONTIERS IN HANDWRITING RECOGNITION (ICFHR) | 2014年

基金：

加拿大自然科学与工程研究理事会; 欧盟第七框架计划;

关键词：

REPRESENTATION;

D O I：

10.1109/ICFHR.2014.97

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A method for automatic generation of synthetic handwritten words is presented which is based in the Kinematic Theory and its Sigma-lognormal model. To generate a new synthetic sample, first a real word is modelled using the Sigma log-normal model. Then the Sigma-lognormal parameters are randomly perturbed within a range, introducing human-like variations in the sample. Finally, the velocity function is recalculated taking into account the new parameters. The synthetic words are then used as training data for a Hidden Markov Model based on-line handwritten recognizer. The experimental results confirm the great potential of the Kinematic Theory of rapid human movements applied to writer adaptation.

引用

页码：543 / 548

页数：6

共 2 条

[1] Recent developments in the study of rapid human movements with the kinematic theory: Applications to handwriting and signature synthesis
Plamondon, Rejean
O'Reilly, Christian
Galbally, Javier
Almaksour, Abdullah
Anquetil, Eric
PATTERN RECOGNITION LETTERS, 2014, 35 : 225 - 235
[2] Gestures a Go Go: Authoring Synthetic Human-Like Stroke Gestures Using the Kinematic Theory of Rapid Movements
Leiva, Luis A.
Martin-Albo, Daniel
Plamondon, Rejean
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY, 2016, 7 (02)

← 1 →