Python == 3.8 (>=3.8)
CUDA == 11.8 (>=11.2)
PyTorch == 2.0.1 (>=1.10.0)
mmdet3d == 1.0.0rc6
Following mmdetection3D
a. Create a conda virtual environment and activate it.
conda create -n viewformer python=3.8 -y
conda activate viewformerb. Install PyTorch and torchvision following the official instructions.
e. Install mmdet3d.
pip install mmcv-full==1.6.0
pip install mmdet==2.28.2
pip install mmsegmentation==0.30.0
cd ./ViewFormer-Occ
git clone https://github.com/open-mmlab/mmdetection3d.git
cd mmdetection3d
git checkout v1.0.0rc6
pip install -e .f. Compile CUDA operators of DCNv3 (optional: use the InternImage backbone only).
cd ./ops_dcnv3
sh ./make.sh