How to estimate position using images.

I am expected to create an algorithm (it doesn't have to be a machine learning model) to calculate the change in coordinates of a drone, which is observing an urban field below through its camera directed downwards. (Also elevation doesn't change)

The coordinates of the first minute of the video are provided, and I am tasked with determining the coordinates relative to the initial position (where the first frame is considered (0,0)).

Do you have any suggestions on what type …

algorithm change computervision create drone images machine machine learning machine learning model through urban video