We recommend miniconda to create a virtual Python environment and install the required libraries:
conda env create -f environment.yml
1.) Preprocess images with OpenCV using preprocess.py script.
2.) Import image sequence to Fiji and run TrakEM2 plugin for spatial alignment.
Tutorial: https://www.youtube.com/watch?v=kUyXQRcKWOk&t=2s
3.) Export aligned image sequence from TrakEM2 to .tif and convert to .png using convert_tiff.py script.
4.) Build 3D model from aligned .png images using build_model.py.