Our project involves building NeRF priors and utilizing these priors for downstream perception tasks. Each task requires different environments. Let's begin with building the NeRF priors.
Our project is based on Nerfstudio-0.3.3. If you encounter any issues during installation, please refer to the original repository for more details.
a. Create a new environment and activate it:
conda create -n presight python=3.8 -y
conda activate presight
b. Install pytorch. We use pytorch 1.13+cu117:
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
c. Install tiny-cuda-nn and build the torch bindings. Please refer to the official repository for more details.
You can either install via conda and pip:
conda install -c "nvidia/label/cuda-11.7.1" cuda-toolkit
pip install ninja git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch
or by compiling from source:
git clone --recursive https://github.com/nvlabs/tiny-cuda-nn
cd tiny-cuda-nn
cmake . -B build -DCMAKE_BUILD_TYPE=RelWithDebInfo
cmake --build build --config RelWithDebInfo -j
cd bindings/torch
python setup.py install
d. Install our modified version of Nerfstudio:
cd nerfstudio-0.3.3
pip install -e .
We use BEVDet as our occupancy prediction baseline model. Below are the installation instructions. If you encounter any issues, please refer to the original repository for more details.
a. Create a new environment and activate it.
conda create -n bevdet python=3.8 -y
conda activate bevdet
b. Install pytorch. We use pytorch 1.13+cu117:
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
c. Install mmcv series:
pip install mmcv-full==1.5.3 -f https://download.openmmlab.com/mmcv/dist/cu117/torch1.13/index.html
pip install mmdet==2.25.1 mmsegmentation==0.25.0
d. Install BEVDet:
cd occupancy
pip install -v -e .
We use a slightly modified version of StreamMapNet as our online mapping baseline model. Below are the installation instructions. If you encounter any issues, please refer to the original repository for more details.
a. Create a new environment and activate it:
conda create -n streammapnet python=3.8 -y
conda activate streammapnet
b. Install pytorch. We use pytorch 1.13+cu117:
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
c. Install mmcv series:
pip install mmcv==1.7.0 -f https://download.openmmlab.com/mmcv/dist/cu117/torch1.13/index.html
pip install mmdet==2.28.2
pip install mmsegmentation==0.30.0
pip install mmdet3d==1.0.0rc6
d. Install other requirements:
cd online-mapping
pip install -r requirements.txt