TY  - JOUR
AU  - Volkmar Frinken
AU  - Andreas Fischer
AU  - Markus Baumgartner
AU  - Horst Bunke
PY  - 2014//
TI  - Keyword spotting for self-training of BLSTM NN based handwriting recognition systems
T2  - PR
JO  - Pattern Recognition
SP  - 1073
EP  - 1082
VL  - 47
IS  - 3
KW  - Document retrieval
KW  - Keyword spotting
KW  - Handwriting recognition
KW  - Neural networks
KW  - Semi-supervised learning
N2  - The automatic transcription of unconstrained continuous handwritten text requires well trained recognition systems. The semi-supervised paradigm introduces the concept of not only using labeled data but also unlabeled data in the learning process. Unlabeled data can be gathered at little or not cost. Hence it has the potential to reduce the need for labeling training data, a tedious and costly process. Given a weak initial recognizer trained on labeled data, self-training can be used to recognize unlabeled data and add words that were recognized with high confidence to the training set for re-training. This process is not trivial and requires great care as far as selecting the elements that are to be added to the training set is concerned. In this paper, we propose to use a bidirectional long short-term memory neural network handwritten recognition system for keyword spotting in order to select new elements. A set of experiments shows the high potential of self-training for bootstrapping handwriting recognition systems, both for modern and historical handwritings, and demonstrate the benefits of using keyword spotting over previously published self-training schemes.
L1  - http://refbase.cvc.uab.es/files/FFB2014.pdf
UR  - http://dx.doi.org/10.1016/j.patcog.2013.06.030
N1  - DAG; 600.077; 602.101
ID  - Volkmar Frinken2014
ER  -