Our implementation is based on mmdetection3d
- mmcv-full==1.3.17
- mmdet==2.19.0
- mmsegmentation==0.20.0
- mmdet3d==0.18.1
Following the instruction: https://github.com/open-mmlab/mmdetection3d/blob/v0.18.1/docs/en/getting_started.md
** a. Create a conda virtual environment and activate it.**
conda create -n unibev python=3.7 -y
conda activate unibev
b. Install PyTorch and torchvision following the official instructions.
conda install pytorch==1.10.0 torchvision==0.11.0 torchaudio==0.10.0 cudatoolkit=11.3 -c pytorch -c conda-forge
c. Install MMCV
pip install mmcv-full==1.3.17 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.10.0/index.html
d. Install MMDetection
pip install mmdet==2.19.0
e. Install MMSegmentation
pip install mmsegmentation==0.20.0
f. Clone the MMDetection3D repository
git clone https://github.com/open-mmlab/mmdetection3d.git $MMDET3D_ROOT
cd mmdetection3d
git checkout tags/v0.18.1
g. Integrate our project into mmdetection3d
ln -s $UniBEV_ROOT/projects/UniBEV $MMDET3D_ROOT/projects
ln -s $UniBEV_ROOT/projects/UniBEV/unibev_plugin $MMDET3D_ROOT/mmdet3d
h. Install build requirements and then install MMDetection3D
pip install -v -e . # or "python setup.py develop"
i. Install some extra packages to avoid some bugs
pip install yapf==0.40.1
conda install setuptools=58.2.0