Setting up AutoSfM

Setting up Docker

...

Upon succesful build, an image named 'sfm' will be created on the host. Verify that the image is built using:

...

Minicinda is a barebones Anaconda installation, which only installs the conda (package management) functionality. This is prefereable to the full Anaconda, which installs many data science packages, and is very bulky. To install miniconda, refer to the official guide: https://docs.conda.io/projects/conda/en/latest/user-guide/install/linux.html . Note that the exact name of the executable will vary by the version of Miniconda used.

mkdir ~/software
# Copy the Miniconda installer to software
cd ~/software
bash Miniconda3-py39_4.12.0-Linux-x86_64.sh

When prompted to accept the agreement, type “yes” yes and press Enter

...

When prompted for the installation location, press Enter. This will keep the default path

...

When prompted, type “yes” yes and press Enter

...

Login again and conda will be active.

...

cd ~
git clone -b develop https://github.com/precision-sustainable-ag/SemiF-AnnotationPipeline.git

The repository and the code is now set up. Note that the -b develop is a temporary measure till the main branch is updated with the stable changes.

Setting up the conda environment

Make a new conda environment from the environment.yml file in the repository

# OpenCV dependencies
sudo apt-get install ffmpeg libsm6 libxext6  -y
cd ~/SemiF-AnnotationPipeline

Create GPU or NoGPU environment

No Nvidia GPU:

conda env create -f environment_nogpu.yml

Nvidia GPU:

conda env create -f environment_gpu.yml

Then activate the conda environment

conda activate semif_nogpu

Copy the metashape software to autosfm/volumes/metashape. The installation will look like this:

...