File size: 2,571 Bytes
9e96356 5bfba9e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
# GenerateCT: Text-Guided 3D Chest CT Generation
Welcome to the official repository of GenerateCT, a pioneering work in text-conditional 3D medical image generation, with a particular focus on chest CT volumes. Here, you will find the code and all models for text-to-CT generation, all freely accessible for researchers.
<p align="center">
<img src="figures/examples.gif" width="100%">
</p>
## Results
Our models achieve the following performances :
| | Resolution | Dimension | Text-Guided | FID (↓) | FVD (↓) | CLIP (↑) |
|--------------------|-----------------|------------------|------------------|------------------------------|-----------|-------------|
| CT-ViT | 128 | 3D | No | 73.4 | 1817.4 | N/A |
| Transformer | 128 | 3D | Yes | 104.3 | 1886.8 | 25.2 |
| Diffusion | 512 | 2D | Yes | 14.9 | 409.8 | 27.6 |
| **GenerateCT** | 512 | 3D | Yes | 55.8 | 1092.3 | 27.1 |
## License
Our work, including the codes, trained models, and generated data, is released under a [Creative Commons Attribution (CC-BY) license](https://creativecommons.org/licenses/by/4.0/). This means that anyone is free to share (copy and redistribute the material in any medium or format) and adapt (remix, transform, and build upon the material) for any purpose, even commercially, as long as appropriate credit is given, a link to the license is provided, and any changes that were made are indicated. This aligns with our goal of facilitating progress in the field by providing a resource for researchers to build upon.
## Acknowledgements
We would like to express our gratitude to the following repositories for their invaluable contributions to our work: [Phenaki Pytorch by Lucidrains](https://github.com/lucidrains/phenaki-pytorch), [Phenaki by LAION-AI](https://github.com/LAION-AI/phenaki), [Imagen Pytorch by Lucidrains](https://github.com/lucidrains/imagen-pytorch), and [CT Net Models by Rachellea](https://github.com/rachellea/ct-net-models). We extend our sincere appreciation to these researchers for their exceptional open-source efforts. If you utilize our models and code, we kindly request that you also consider citing these works to acknowledge their contributions.
|