README.md

---
license: apache-2.0
language:
- en
pipeline_tag: text-to-image
tags:
- mamba
- generative model
- stable diffusion
- stochastic interpolant
- zigma
- zigzag

---
# ZigMa: A DiT-style Zigzag Mamba Diffusion Model (ECCV 2024)

This model represents the official checkpoint of the paper titled ["ZigMa: A DiT-style Zigzag Mamba Diffusion Model"](https://huggingface.co/papers/2403.13802).


[![Website](doc/badges/badge-website.svg)](https://taohu.me/project_zigma)
[![Paper](https://img.shields.io/badge/arXiv-PDF-b31b1b)](https://arxiv.orgg)
[![Hugging Face Model](https://img.shields.io/badge/🤗%20Hugging%20Face-Model-green)](https://huggingface.co/Bingxin/Marigold)
[![License](https://img.shields.io/badge/License-Apache--2.0-929292)](https://www.apache.org/licenses/LICENSE-2.0)


[Vincent Tao Hu](http://taohu.me),
[Stefan Andreas Baumann](https://stefan-baumann.eu/),
[Ming Gui](https://www.linkedin.com/in/ming-gui-87b76a16b/?originalSubdomain=de),
[Olga Grebenkova](https://www.linkedin.com/in/grebenkovao/),
[Pingchuan Ma](https://www.linkedin.com/in/pingchuan-ma-492543156/),
[Johannes Fischer](https://www.linkedin.com/in/js-fischer/ )
[Bjorn Ommer](https://ommer-lab.com/people/ommer/ )

We present ZigMa, a scanning scheme that follows a zigzag pattern, considering both spatial continuity and parameter efficiency. We further adapt this scheme to video, separating the reasoning between spatial and temporal dimensions, thus achieving efficient parameter utilization. Our design allows for greater incorporation of inductive bias for non-1D data and improves parameter efficiency in diffusion models.

![teaser](doc/teaser_3col.png)



## 🎓 Citation

```bibtex
@InProceedings{hu2024zigma,
      title={ZigMa: A DiT-style Zigzag Mamba Diffusion Model},
      author={Vincent Tao Hu and Stefan Andreas Baumann and Ming Gui and Olga Grebenkova and Pingchuan Ma and Johannes Fischer and Bjorn Ommer},
      booktitle = {Arxiv},
      year={2024}
}
```

## 🎫 License

This work is licensed under the Apache License, Version 2.0 (as defined in the [LICENSE](LICENSE.txt)).

By downloading and using the code and model you agree to the terms in the  [LICENSE](LICENSE.txt).

[![License](https://img.shields.io/badge/License-Apache--2.0-929292)](https://www.apache.org/licenses/LICENSE-2.0)