Given a scholarly PDF, extract figures, tables, captions, and section titles. - View it on GitHub