Fitting an ellipsoid to 3D data points

Question

I have a large set of 3D data points to which I want to fit to an ellipsoid.

My maths is pretty poor, so I\'m having trouble implementing the least squares method withou

Answer 1

Here is very simple method to find foci's, based on random search. No linear algebra used. So if you are fine with non-perfect solution:

3DPoints - Array of some Amount of points
vecCenter - average of 3DPoints
AngleCosine - cos of angle between two vectors

RandomOrder(3DPoints)

vecFocus := (0, 0, 0)
for i := 0 to Amount:
    vecRadius := 3DPoints[i] - vecCenter

    // Change vecRadius direction to parallel
    if AngleCosine(vecFocus, vecRadius) < 0 then
        vecRadius *= -1
    vecFocus += (vecRadius - vecFocus) / Amount

Result can be improved by passing array 2nd time in backward direction Foci's placed at vecCenter +/- vecFocus