This chrome extension automatically parses every image with computer vision algorithms in order to extract textual regions and allow interaction- highlighting and copying just like ordinary text.
Images as Text
Highlight and copy-paste text in pictures as if they were just ordinary text on a web page
Updates
Here's what's happened in the past 5 years:
This project eventually evolved into Project Naptha (https://projectnaptha.com/) which has ~170,000 active users. Now, in addition to merely selecting and highlighting text, you can also use it to translate and erase text from images. It's been featured on Hacker News, Engadget, The Verge, Lifehacker, and ZDNet.
In the process we've spun off a few key components as open source libraries— most notably Tesseract.JS (https://github.com/naptha/tesseract.js) an optical character recognition library for Javascript with over 17,000 Github stars.
Looking back— I'm grateful for everyone who's helped along the way. To Jerome Wu for taking over the TesseractJS project. To Guillermo Webster, who woke me up when the hackathon judges walked by. And most of all to everyone who's downloaded and installed the app.
Leave feedback in the comments!
Log in or sign up for Devpost to join the conversation.
Log in or sign up for Devpost to join the conversation.