Google Chrome above all other popular web browsers is known for its extensions. These include facebook zoom, grooveshark and disctionaries. More innovative ones include 100,000 stars (a view of the nearest 100,000 stars in the milky way galaxy) and a view of J.R.R Tolkien’s Middle Earth. Recently another extension called Project Naptha has been added to that list.
Naptha was coded by developer Kevin Kwok. It allows the user to select any text in a certain picture and manipulate it in a lot of ways. The text can be copied, erased, translated and edited depending on the user’s choice. It recognizes text from any picture using both Microsoft and Google software.
The feature works well because of an inbuilt technique called in-painting, the effect of which is similar to that of Adobe Photoshop’s “Content-aware-fill” feature. In-painting adjusts the translated, edited, erased or modified text such that the picture does not seem altered by filling in the new areas that need recoloring with colors from the surrounding areas in the picture. And of course no editing software is complete without a save feature allowing you to reminisce later in life about your handiwork, no matter how simple or silly it might be.
The extension runs on the Stroke Width Transform algorithm Microsoft Research invented for text detection in natural scenes. It also provides the option of using Google’s open-source OCR engine Tesseract when necessary.
Project Naptha will soon be coming to other browsers like Internet Explorer, Opera, Firefox and Safari if the developer Mr. Kwok is to be believed.