Full-Page Text Recognition: Learning Where to Start and When to Stop
This addresses text recognition in real-life documents, but it is incremental as it builds on existing localization methods with a specific efficiency improvement.
The paper tackles the problem of text line detection and localization in heterogeneous documents by predicting only the left side of text lines and letting the recognizer determine the end, achieving good results on the Maurdor dataset.
Text line detection and localization is a crucial step for full page document analysis, but still suffers from heterogeneity of real life documents. In this paper, we present a new approach for full page text recognition. Localization of the text lines is based on regressions with Fully Convolutional Neural Networks and Multidimensional Long Short-Term Memory as contextual layers. In order to increase the efficiency of this localization method, only the position of the left side of the text lines are predicted. The text recognizer is then in charge of predicting the end of the text to recognize. This method has shown good results for full page text recognition on the highly heterogeneous Maurdor dataset.