Memeify

This repository provides parts of the code and datasets for the paper - Memeify: A Large Scale Meme Generation System published at CODS-COMAD 2020. The code has been tested on Python 3.6.8 and Tensorflow 1.14.

Datasets

The meme captions and images are made available here.

Generation

The code for generating and sampling new memes can be found in the Memeify.ipynb notebook. For varying the quality of memes, please tune the temperature parameter in the code to modify the softmax distribution behaviour.

Demonstration

A demo video of our paper is available here.

Acknowledgements

Parts of the code have been used from this repository.

Citation

If you found our work useful, please consider citing it as:

@incollection{vyalla2020memeify,
  title={Memeify: A Large-Scale Meme Generation System},
  author={Vyalla, Suryatej Reddy and Udandarao, Vishaal},
  booktitle={Proceedings of the 7th ACM IKDD CoDS and 25th COMAD},
  pages={307--311},
  year={2020}
}

In case of any queries, you can contact us: