I’m trying to figure out if there’s a way to determine the extrinsic camera matrix/distortion coefficients for a camera I no longer have access to. It’s a bit of a unique situation.

I can’t perform the chessboard calibration using OpenCV because I don’t have the camera anymore. However, I do have thousands of photos taken with it, each featuring a different object. I also know the camera model and the focal length used for each shot. Is it possible to reverse engineer the extrinsic matrix given these thousands of photos?

What I am currently thinking of doing is finding objects that I know are straight lined and then fitting a polynomial to these lines and then try and optimize from here? Any help is appreciated and please let me know if this is even possible to figure out.