It’s called Project Naptha and can recognize, edit and translate into other languages the words that appear in any picture; is as OCR tools, but integrated into the browser.
For a long time how to digitalize a text depended on taking a photo (with a scanner, usually) and use a tool optical image recognition (OCR) to convert text on paper into digital content. But now the computer does a step: transform an image into text characters manipulable by computer, without further complication to install an extension in the browser.
In one of the hackathon regularly organized by MIT, Kevin Kwok, a student of the university, presented the Project Naptha, a tool that, through an algorithm, can recognize and copy any text within an image.
Unlike traditional OCR applications, does not require using complementary software or upload the images to a site; everything is done within the browser and automatically, and the translation is instant.
The tool, still in development, functions as a free browser extension for Chrome and can be downloaded from the site projectnaptha.com. Once installed, on the same site there are different images that work as a demo to test its effectiveness.
The application works both with the shortcut keys for your browser such as context menu, which offers advanced options to copy, edit, translate and correct the spelling of the text.