Installation

Our code is tested on the following environment.

1. Create conda environment

conda create -n selfocc python=3.8.16
conda activate selfocc

2. Install PyTorch

pip install torch==2.0.0 torchvision==0.15.1 torchaudio==2.0.1 --index-url https://download.pytorch.org/whl/cu118

3. Install packages from MMLab

pip install openmim
mim install mmcv==2.0.1
mim install mmdet==3.0.0
mim install mmsegmentation==1.0.0
mim install mmdet3d==1.1.1

4. Install other packages

pip install spconv-cu117
pip install timm
pip install SharedArray
pip install easydict
pip install git+https://github.com/NVIDIA/gpu_affinity

4. Install custom CUDA ops

cd model/encoder/gaussian_encoder/ops && pip install -e .
cd model/head/localagg && pip install -e .
python model/ops/setup.py develop

5. (Optional) For visualization

pip install pyvirtualdisplay mayavi matplotlib==3.7.2 PyQt5