install.md

Step-by-step installation instructions

Following https://mmdetection3d.readthedocs.io/en/latest/getting_started.html#installation

a. Create a conda virtual environment and activate it.

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

b. Install PyTorch and torchvision following the official instructions.

pip install torch==1.9.1+cu111 torchvision==0.10.1+cu111 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html
# Recommended torch>=1.9

c. Install gcc>=5 in conda env (optional).

conda install -c omgarcia gcc-5 # gcc-6.2

d. Install mmcv-full.

pip install mmcv-full==1.4.0
#  pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html

e. Install mmdet and mmseg.

pip install mmdet==2.14.0
pip install mmsegmentation==0.14.1

f. Clone VMA.

git clone https://github.com/hustvl/VMA.git

g. Install mmdet3d

cd /path/to/VMA/mmdetection3d
python setup.py develop

h. Install other requirements.

cd /path/to/VMA
pip install -r requirements.txt

i. Prepare pretrained models.

We use iCurb backbone provided in iCurb to train NYC Planimetric Database and resnet-152 to train our custom data.

cd /path/to/VMA
mkdir ckpts

cd ckpts 
wget https://download.pytorch.org/models/resnet152-b121ed2d.pth

You can download our remapped iCurb backbone in Baidu/Google.