Skip to content

Install ROMI software for virtual plants acquisition & reconstructionLink

To follows this guide you should have a conda or a Python venv, see here.

For the sake of clarity the environment will be called virtual_plants.

Notice for using the virtual scannerLink

If you want to use the virtual scanner, the modified python version bundled with blender and the environment python version have to match.

To obtain the python version bundled with your distribution of blender, type:

blender -b --python-expr "import sys; print(sys.version)"

It will output something like:

Blender 2.82 (sub 7) (hash 5b416ffb848e built 2020-02-14 16:19:45)
ALSA lib pcm_dmix.c:1089:(snd_pcm_dmix_open) unable to open slave
3.8.1 (default, Jan 22 2020, 06:38:00) 
[GCC 9.2.0]

Blender quit

In this case, this means Blender bundle Python 3.8, and you should too.

In the following, we will assume that you are using conda environments. If not, adapt with corresponding virtualenv commands.

Install ROMI packages with pip:Link

Activate your virtual_plants environment!


Since this is still under development, the packages are installed in "editable mode" with the -e option.

Install openalea.lpyLink

If you're using python>=3.7 and conda, just install lpy from conda:

conda install -c conda-forge -c fredboudon openalea.lpy

Install romicgal sourcesLink

To pilot the hardware you have to install the plantimagerpackage:

python3 -m pip install -e git+


This takes some time since it has to download dependencies (CGAL-0.5 & boost-1.72.0) and compile them.

Install plant3dvision sourcesLink

To start "acquisition jobs", you have to install the plant3dvision package:

python3 -m pip install -e git+

Install plantdb sourcesLink

Since we will need an active database to export the acquisitions, you have to install the plantdb package:

python3 -m pip install -e git+

Install romiseg sourcesLink

To install the additional segmentation module:

python3 -m pip install git+


If not using CUDA 10.*, you have to install the matching pytorch distribution. For example, for CUDA 9.2, use:

pip install torch==1.4.0+cu92 -f

Example databaseLink

To quickly create an example DB you can use:

tar -xf test_db.tar.gz

This will create a integration_tests folder with a ready to use test database.

You should now be ready to perform tasks on virtual plants such as "creation", "acquisition" & "reconstruction" following the dedicated user guide.