Page Layout Analysis System for Unconstrained Historic Documents
This work addresses the challenge of automatic transcription for unconstrained historic documents, which is an incremental improvement over existing methods.
The authors tackled the problem of extracting text regions and lines from historic documents by extending a CNN-based baseline detection system with line height, text block boundary, and text orientation predictions, demonstrating strong performance on the cBAD baseline detection dataset and benchmarking on their new public PERO layout dataset.
Extraction of text regions and individual text lines from historic documents is necessary for automatic transcription. We propose extending a CNN-based text baseline detection system by adding line height and text block boundary predictions to the model output, allowing the system to extract more comprehensive layout information. We also show that pixel-wise text orientation prediction can be used for processing documents with multiple text orientations. We demonstrate that the proposed method performs well on the cBAD baseline detection dataset. Additionally, we benchmark the method on newly introduced PERO layout dataset which we also make public.