Bidirectional Decoder Networks for Attention-Based End-to-End Offline Handwriting Recognition
Authors:
- Patrick DoetschPDPatrick Doetsch
- Albert ZeyerAZAlbert Zeyer
- Hermann NeyHNHermann Ney
- 1 October 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21676445,p. 361-366
Abstract
Recurrent neural networks that can be trained end-to-end on sequence learning tasks provide promising benefits over traditional recognition systems. In this paper, we demonstrate the application of an attention-based long short-term memory decoder network for offline handwriting recognition and analyze the segmentation, classification and decoding errors produced by the model. We further extend the decoding network by a bidirectional topology together with an integrated length estimation procedure and show that it is superior to unidirectional decoder networks. Results are presented for the word and text line recognition tasks of the RIMES handwriting recognition database. The software used in the experiments is freely available for academic research purposes.Keywords
This publication has 10 references indexed in Scilit:
- Recursive Recurrent Nets with Attention Modeling for OCR in the Wild2016
- Segmentation-free handwritten Chinese text recognition with LSTM-RNN2015
- Recurrent neural network language model training with noise contrastive estimation for speech recognition2015
- Fast and Robust Training of Recurrent Neural Networks for Offline Handwriting Recognition2014
- Hybrid speech recognition with Deep Bidirectional LSTM2013
- The A2iA French handwriting recognition system at the Rimes-ICDAR2011 competition2012
- ICDAR 2011 - French Handwriting Recognition Competition2011
- Connectionist temporal classification2006
- Bidirectional recurrent neural networksIEEE Transactions on Signal Processing, 1997
- Long Short-Term MemoryNeural Computation, 1997