Find answers from the community

Updated 9 months ago

Hi everyone!

Hi everyone!
Does anyone have experience in extracting images from PDFs? I’m working with math PDF courses and some illustrations are interesting for other parts of my processes. I’ve tried generating them on the fly, low quality, and I’ve tried generating the code that generates shapes or curves (matplotlib) and it’s kinda hit or miss.

4 comments

TTeemu

You can use a library like PyMuPDF to extract all the images

FFares

I only found this experiment by llama index: https://docs.llamaindex.ai/en/stable/examples/multi_modal/multi_modal_pdf_tables.html

FFares

that implies that you need to convert every page to an image, then run multimodal queries

LLogan M

Have you tried llama-parse?
https://github.com/run-llama/llama_parse/blob/main/examples/demo_json.ipynb

Add a reply