1

I am trying to understand nonlinear least squares to fit spheres in point cloud data. I kind of understand how the linear least squares works but every kind of paper on nonlinear least squares/Levenberge-Marquerdt is just pure magic to me. Well, it's more that I just don't understand that level of math.

I know that you can linearize(?) the sphere function to use linear least squares but the PCL implementation of a sphere fit uses Levenberge-Marquerdt, which is why I have to understand what I am using, if that makes sense.

If possible, can anyone make a small example on how to use the Levenberge-Marquerdt algorithm?

Royi
  • 10,050
Nope
  • 11
  • Do you mean the Levenberg-Marquardt algorithm described at https://en.wikipedia.org/wiki/Levenberg%E2%80%93Marquardt_algorithm ? – kimchi lover Aug 28 '22 at 20:13
  • @kimchilover Honest to god, I barely understand what the steps are exactly. "Standard" Least Squares is explained so much easier. I already read the wiki like 15 times now lol – Nope Aug 28 '22 at 23:27

0 Answers0