LayoutReader: Pre-training of Text and Layout for Reading Order Detection
This addresses the lack of large annotated datasets for reading order detection, enabling deep learning applications in document understanding, though it is incremental in leveraging existing document metadata.
The paper tackled the problem of reading order detection in visually-rich documents by constructing ReadingBank, a large-scale dataset of 500,000 document images with automated annotations, and proposed LayoutReader, a seq2seq model that achieves near-perfect detection and significantly improves OCR engines.
Reading order detection is the cornerstone to understanding visually-rich documents (e.g., receipts and forms). Unfortunately, no existing work took advantage of advanced deep learning models because it is too laborious to annotate a large enough dataset. We observe that the reading order of WORD documents is embedded in their XML metadata; meanwhile, it is easy to convert WORD documents to PDFs or images. Therefore, in an automated manner, we construct ReadingBank, a benchmark dataset that contains reading order, text, and layout information for 500,000 document images covering a wide spectrum of document types. This first-ever large-scale dataset unleashes the power of deep neural networks for reading order detection. Specifically, our proposed LayoutReader captures the text and layout information for reading order prediction using the seq2seq model. It performs almost perfectly in reading order detection and significantly improves both open-source and commercial OCR engines in ordering text lines in their results in our experiments. We will release the dataset and model at \url{https://aka.ms/layoutreader}.