This repository is to showcase my experience in the field of medical visualization as a research assistant under the supervision of Prof. Dr. Sylvia Saalfeld.
As the images had muti-level resolution (e.g., level 0 max resolution), I have processed the level 5 for all the images taking the memory constraint of my workspace environment into consideration. I have used both register_transform (one pair at a time) and register_transform_stack (co-registers a stack of frames) from python. For each, I have started from the middle index of the dataset and registered towards both the end and the start. For the co-registration, I used both strategies: pair by pair ( every time the latest registered image becomes the reference image) and with single reference image ( the middle indexed image remains constant as the reference image). For resizing, the dimensions of the reference image was applied on the offset image in all cases.