README.md

---
license: apache-2.0
---
# DATID-3D: Diversity-Preserved Domain Adaptation Using Text-to-Image Diffusion for 3D Generative Model
[Gwanghyun Kim](https://gwang-kim.github.io/), [Se Young Chun](https://icl.snu.ac.kr/pi) <br>
CVPR 2023 <br>
[gwang-kim.github.io/datid_3d](gwang-kim.github.io/datid_3d/)


We propose DATID-3D, a novel pipeline of text-guided domain adaptation tailored for 3D generative models using text-to-image diffusion models that can synthesize diverse images per text prompt without collecting additional images and camera information for the target domain.** Unlike 3D extensions of prior text-guided domain adaptation methods, our novel pipeline was able to fine-tune the state-of-the-art 3D generator of the source domain to synthesize high resolution, multi-view consistent images in text-guided targeted domains without additional data, outperforming the existing text-guided domain adaptation methods in diversity and text-image correspondence. Furthermore, we propose and demonstrate diverse 3D image manipulations such as one-shot instance-selected adaptation and single-view manipulated 3D reconstruction to fully enjoy diversity in text.

## Fine-tuned 3D generative models

Fine-tuned 3D generative models using DATID-3D pipeline are stored as `*.pkl` files.
You can download the models in [our Hugginface model pages](https://huggingface.co/gwang-kim/datid3d-finetuned-eg3d-models/tree/main/finetuned_models).


## Citation

```
@inproceedings{kim2022datid3d,
  author = {DATID-3D: Diversity-Preserved Domain Adaptation Using Text-to-Image Diffusion for 3D Generative Model},
  title = {Gwanghyun Kim and Se Young Chun},
  booktitle = {CVPR},
  year = {2023}
}
```

========================================================