I didn’t even realize how long I had been waiting for a tool like this! Adam Kucharski (https://bsky.app/profile/adamjkucharski.bsky.social), professor of Epidemiology and Mathematics in London, developed a simple web tool to extract numerical data directly from plots embedded in PDFs. Extremely useful for, e.g., revisiting published figures when raw data are unavailable.
🌍 https://adamkucharski.github.io/pdf2plot/
📼 by Adam Kucharski
@FabMusacchio indeed, it's much better than the previous tools I've been using, though I'm not curious whether it would work on all images and whether you can work around the issues when automatic recognition breaks. But when it works, it's really a great time saver!
Also, someone has been asking me for the same concept but to extract data from tables in PDFs or images, in case you know a similarly great tool...