ubc-vision / Make-A-Story

Code Release for the paper "Make-A-Story: Visual Memory Conditioned Consistent Story Generation" in CVPR 2023

We are still cleaning our Code

Make-A-Story

This repository contains the code for the CVPR 2023 paper titled "Make-A-Story: Visual Memory Conditioned Consistent Story Generation".

Extended dataset

We extend existing datasets (e.g. "Mugen", "FlintstonesSV" and "PororoSV" dataset) to include more complex scenarios and referential text.

Acknowledgment

This repository is developed on top of "Latent Diffusion Models". Please also refer to the original License of the project.

Bibtext

If you find this code is useful for your research, please cite our paper

@article{rahman2022make,
  title={Make-A-Story: Visual Memory Conditioned Consistent Story Generation},
  author={Rahman, Tanzila and Lee, Hsin-Ying and Ren, Jian and Tulyakov, Sergey and Mahajan, Shweta and Sigal, Leonid},
  journal={arXiv preprint arXiv:2211.13319},
  year={2022}
}