You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Mhh, I think the same thing is possible in 3D, except that only the second derivatives are known (not the first derivatives). This makes the implementation considerably more sketchy, as we need to extract the first derivative using a finite difference method. The accuracy gains would then not be as obvious.
To properly implement this, we need a validation showing us:
The interpolation accuracy, as a function of the number of samples along the optical axis
The accuracy for a regular CubicSpline interpolation
The accuracy once we use the information about the second derivatives (using a quintic spline interpolation)
Since this is quite a bit of work, and the ROI is not as obvious, moving this to longterm
The interpolation technique used to speed up ray tracing can be improved
The text was updated successfully, but these errors were encountered: