More and more students are catching what is called, the meme virus. Students with it lie in bed days on end tagging their friends on social media such as Facebook. It's time to tackle this problem head on.

What it does

MemeBlocker+ uses a trained neural net to classify scraped images on the current webpage into either 'Meme' or 'Not-Meme'. If the image falls under the category of meme, then it deletes it from the displayed html.

How we built it

We first built a database filled with thousands of memes and non-meme photos as our two categories. Next we trained our neural net using this database.

Challenges we ran into

Accomplishments that we're proud of

Using machine learning and helping reduce the meme virus.

What we learned

What's next for MemeBlocker+

MemeBlocker+ can be expanded to replace non-meme pictures with memes if wanted using SIFT. A user could potentially ask for meme (or non-meme) images to be replaced with images belonging to a tagword of their choice.

Built With

Share this project: