Convert PDFs and images to structured text and layout data
Read text from images using scene text recognition models
