conda create -n MVIPnerf python=3.8
conda activate MVIPnerf
pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu117
pip install -r requirements.txt
pip install -r requirements_df.txt
pip install lpips
pip install ConfigArgParse

1
├── images
│   ├── IMG_2707.jpg
│   ├── IMG_2708.jpg
│   ├── ...
│   └── IMG_2736.jpg
└── images_4
    ├── IMG_2707.png
    ├── IMG_2708.png
    ├── ...
    ├── IMG_2736.png
    └── label
        ├── IMG_2707.png
        ├── IMG_2708.png
        ├── ...
        └── IMG_2736.png
    └── Depth
        ├── IMG_2707.png
        ├── IMG_2708.png
        ├── ...
        └── IMG_2736.png

python DS_NeRF/run.py --config DS_NeRF/config/config_1.txt

datadir: folder for the dataset
factor: downscale of the image resolution of the inpainted scene 
is_normal_guidance: control whether using normal guidance
is_colla_guidance: control whether using multi-view guidance
text: text prompt for the inpainted scene
normalmap_render_factor: we use a factor to downscale the rendered normal map, due to the RAM limitation

@inproceedings{MVIPNeRF,
      title={MVIP-NeRF: Multi-view 3D Inpainting on NeRF Scenes via Diffusion Prior}, 
      author={Honghua Chen and Chen Change Loy and Xingang Pan},
      year={2024},
      booktitle={CVPR},
}