Step-by-step installation instructions

a. Create a conda virtual environment and activate it.

conda create -n OccCylindrical python=3.8 -y
conda activate OccCylindrical

b. Install PyTorch and torchvision following the official instructions.

conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia

c. Install mmengine, mmcv, mmdet, mmdet3d, and mmseg.

pip install -U openmim
mim install mmengine
mim install 'mmcv>=2.0.0rc4'
mim install 'mmdet>=3.0.0'
mim install "mmdet3d>=1.1.0"
pip install "mmsegmentation>=1.0.0"

e. Download code and backbone pretrain weight.

git clone https://github.com/DanielMing123/OccCylindrical.git
cd OccCylindrical
mkdir ckpt
wget https://github.com/zhiqi-li/storage/releases/download/v1.0/r101_dcn_fcos3d_pretrain.pth

f. The OccCylindrical repo core structure should be like the following

OccCylindrical
├── ckpt/
├── configs/
├── occcylindrical/
├── tools/
├── data/
│   ├── nuscenes/
│   ├── nuScenes-c-image/
│   ├── nuScenes-c-lidar/