Also neat is that the enabling libraries here – Tesseract.js and PDF.js – are both pretty old at this point: First commit to Tesseract.js was Jun 26, 2015 https://
github.com/naptha/tessera
ct.js/commit/906ce3cadbffaf5f7317a4418f282c4b78bf8385
… First to PDF.js was Apr 25, 2011 https://
github.com/mozilla/pdf.js
/commit/6dc1770bba7a417ce5664c0305469e5bb7ea76bd
…
@simonw
-
Ancient Open Source Libraries Power Modern AI Applications
By
–
-
Tesseract.js and PDF.js: Decade-Old Libraries Still Relevant
By
–
I love that both the libraries I'm using here – Tesseract.js and PDF.js – are nearly ten years old!
-
Minimalist 226-Line Tool with PDF and OCR Capabilities
By
–
Something I really like about this tool is that the entire thing is 226 lines of combined HTML, CSS and JavaScript (plus the PDF.js and Tesseract.js dependencies, loaded from a CDN) The code is a little untidy but at 226 lines it honestly doesn't matter
-
Experimenting with Claude 3 and Gemini Pro 1.5 Vision
By
–
Worth a try! I've been experimenting with Claude 3 and Gemini Pro 1.5 vision recently too, they're very powerful
-
Claude 3 Gemini Pro GPT-4 Vision outperform Tesseract OCR
By
–
No, I don't think Tesseract isn't powerful enough for that unfortunately – that's where Claude 3 / Gemini Pro / GPT-4 Vision are likely to be a better fit
-
Tesseract OCR limitations with illustrations and typeset text
By
–
Yeah, that's pretty much expected – Tesseract is great at regular typeset text, but it's really not very effective at illustrations
-
Mastodon DDoS Attack Overwhelms Cloudflare Protection
By
–
It was the Mastodon effect – when you get a DDoS from every Mastodon server when you toot something. Usually Cloudflare protects me but it didn't here for some reason, notes on the incident here
-
OCR Tool Built with Claude 3 Opus and GPT-4
By
–
Try it out here: https://
tools.simonwillison.net/ocr I wrote about how I built it – including all of the prompts I used through both Claude 3 Opus and a little bit of ChatGPT/GPT-4 – on my blog: -
Browser-based OCR tool using Tesseract.js and PDF.js
By
–
I built a new tool! It's a single page web app that runs OCR against images and PDFs entirely in your browser (no file upload needed) using Tesseract.js and PDF.js
— Simon Willison (@simonw) 30 mars 2024
You can drop files onto it, or you can click to select and open them (which works on Mobile Safari as well) pic.twitter.com/HVeFtEXNMmI built a new tool! It's a single page web app that runs OCR against images and PDFs entirely in your browser (no file upload needed) using Tesseract.js and PDF.js You can drop files onto it, or you can click to select and open them (which works on Mobile Safari as well)
-
Verifying GPT-4 Generated Python Code Against Documentation
By
–
For a recent example,see my note here about checking the Python documentation to confirm what readline.set_startup_hook() does after seeing it in some Python that GPT-4 wrote for me https://
simonwillison.net/2024/Mar/26/ll
m-cmd/#assistance-from-chatgpt
…