Monthly Archives: September 2012

Implementing the Gauss-Newton Algorithm for Sphere Fitting (3 of 3)

The first post in this series made it clear that the major computational and storage costs to implementing the Gauss-Newton algorithm come in computing the matrices and . By carefully avoiding materialization of the Jacobian matrix and the residual vector, … Continue reading

Posted in Mathematics | Tagged , , , , , , , , , , | 4 Comments

Implementing the Gauss-Newton Algorithm for Sphere Fitting (2 of 3)

I sketched a naïve implementation of the Gauss-Newton algorithm for sphere fitting, and provided an implementation in Octave. While it works fine on a laptop, it uses far too much memory to handle more than a handful of calibration samples … Continue reading

Posted in Mathematics | Tagged , , , , , , , | 2 Comments

Implementing the Gauss-Newton Algorithm for Sphere Fitting (1 of 3)

In some calibration strategies we encounter a sphere fitting problem: we need to find the map that does the best job sending the observed data to a unit sphere. This simple problem statement leaves two critical questions unanswered. What family … Continue reading

Posted in Mathematics | Tagged , , , , , , | 2 Comments