mangafy | Devpost

Inspiration

Reading a riveting web comic, you're left with a cliffhanger that will only be resolved with next week's update. Out of desperation, you find the comic in a foreign language (which is ahead by a few chapters) and have to try piecing the context from the images alone. We've been in this predicament and felt this frustration too many times. Sometimes, a manga or comic is only available in its original language, which you can't translate using conventional translation programs because they don't detect text on images. We wanted a way to conveniently translate the dialogue existing within images on a browser, which is usually the case for comics.

What it does

Mangafy is a chrome extension that allows you to select an area on your browser and displays the translation for any text found within the selected area. This way, you can quickly and effortlessly translate manga while you're reading on the go. This works for non-manga-related purposes as well, such as translating images you see while you're browsing the internet.

How we built it

We used a chrome extension to allow the user to select text on their browser, Django to act as a server that communicates with the extension, and Google's Vision API to detect text within images.

Challenges we ran into

We had a hard time making the google chrome extension convert the user's area of selection into an image that our Vision API could process. Most of the instructions on the internet only described how to take an image of the entire viewport.

Accomplishments that we're proud of

We're proud of making a functional application that we believe will be useful in our own and other's lives, and persistently working on our project in small steps until we made something bigger.