MeDM: Mediating Image Diffusion Models for Video-to-Video Translation with Temporal Correspondence Guidance
AAAI 2024
Ernie Chu, Tzuhsuan Huang, Shuo-Yen Lin, Jun-Cheng Chen
Research Center for Information Technology Innovation, Academia Sinica
Project Page | Paper | arXiv | Colab
We use conda to maintain the Python environment
conda env create -f environment.yml The implementation of MeDM is in pipeline_medm.py. We incorporate MeDM into the snapshot of Diffusers at version 0.20.0. To install it, simply use
cd diffusers-0.20.0 pip install . We provide two simple examples on how to use MeDMPipeline in colabs. Remember to skip the colab-specific blocks when running locally.
If you find our work useful, please consider cite this work as
@inproceedings{chu2024medm, title={MeDM: Mediating Image Diffusion Models for Video-to-Video Translation with Temporal Correspondence Guidance}, author={Ernie Chu and Tzuhsuan Huang and Shuo-Yen Lin and Jun-Cheng Chen}, booktitle={Proceedings of the AAAI Conference on Artificial Intelligence}, year={2024} }