Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Joint Line Segmentation and Transcription for End-to-End Handwritten Paragraph Recognition

About

Offline handwriting recognition systems require cropped text line images for both training and recognition. On the one hand, the annotation of position and transcript at line level is costly to obtain. On the other hand, automatic line segmentation algorithms are prone to errors, compromising the subsequent recognition. In this paper, we propose a modification of the popular and efficient multi-dimensional long short-term memory recurrent neural networks (MDLSTM-RNNs) to enable end-to-end processing of handwritten paragraphs. More particularly, we replace the collapse layer transforming the two-dimensional representation into a sequence of predictions by a recurrent version which can recognize one line at a time. In the proposed model, a neural network performs a kind of implicit line segmentation by computing attention weights on the image representation. The experiments on paragraphs of Rimes and IAM database yield results that are competitive with those of networks trained at line level, and constitute a significant step towards end-to-end transcription of full documents.

Th\'eodore Bluche• 2016

Related benchmarks

TaskDatasetResultRank
Handwritten text recognitionIAM (test)
CER5.5
102
Handwritten text recognitionRIMES (test)
CER2.9
15
Handwriting RecognitionIAM page paragraph
CER7.9
6
Paragraph-level Handwriting RecognitionRIMES 2011 (test)
CER2.9
4
Showing 4 of 4 rows

Other info

Follow for update