PubLayNet is a large dataset of document images, of which the layout is annotated with both bounding boxes and polygonal segmentations. The source of the documents is PubMed Central Open Access Subset (commercial use collection). The annotations are automatically generated by matching the PDF format and the XML format of the articles in the PubMed Central Open Access Subset. More details are available in our paper "PubLayNet: largest dataset ever for document layout analysis."
Images and annotations can be downloaded from this Box folder
The annotation files follows the json format of the Object Detection task of MS COCO
A Jupyter notebook is provided to generate the following visualization of the annotations of 20 sample pages.
