Inspiration
We set out at the beginning of the hackathon on an amazing journey to build a program which helps us to find only the dankest of memes.
What it does
It downloads images from certain subreddits and finds similar images. In our testcase we are testing mainly me_irl but it would be trivial to add a larger database, provided enough computing power
How we built it
We used python and the reddit API to download images, and then used image hashing to compare images, finding similar images which we deemed to be spicy af.
Challenges we ran into
Installing packages, integrating code. Implementing a hashing system which would show similar images without too much error.
Accomplishments that we're proud of
We can automate finding memes with a similar format.
What we learned
We learnt how to compare images using image hashing, the reddit API, using python and the difficulties of Qt.
What's next for \ue420\ue420\ue420
Implement methods for finding memes that arent just a visually similar format, ie. using text from the image and finding patterns in that. Another advancement would be to find making sure that we can find spicy memes only from a wide variety of places, this would be done by rating the sources of the memes and tracking their quality over time so that we could invest in the meme market and make our millions.
Log in or sign up for Devpost to join the conversation.