Docling only detects images explicitly inserted as PictureItem in PPTX files. It doesn’t extract images embedded in placeholders or grouped shapes because its PPTX parser lacks configurable options like the PDF pipeline (e.g., pipeline_options.generate_picture_images = True). Is there any way to extract such embedded images from PPTX using Docling only,please let me know? I tested with pdf, pptx file, its extracting all images from pdf but not from pptx. Continue reading...
