Spline interpolation of 3D scattered data

Hi, my data consists of two sets of mesh vertices (scattered distributed, each given by three coordinates x,y,z) and a set of values which belongs to the smaller set of vertices. The smaller set of vertices is a subset of the bigger set. Now I want to interpolate the values using the coordinates of the bigger set of vertices. griddata() only supports linear and nearest neighbor interpolation for 3D scattered data. Is there a function in MATLAB to perform a spline interpolation in 3D?

[Matt J]

I'm not sure how you would even define spline interpolation for scattered data. Are you sure you're not talking about a smoothing spline or do you want something that is actually guaranteed to pass through all your given data points?

[Image Analyst]

A screenshot might really help here. Can you upload something?

[Eli Duenisch]

Hi, sorry for the delayed response to your answers. The thing is I use a FEM software (Comsol) to obtain a certain solution vector (which are values assigned to the mesh vertices) for a given mesh. Now I refine that mesh and obtain another solution vector for that one. Then I want to compare the solutions to check for convergence when the mesh gets finer. Because the two solutions do not have the same size I need to increase the smaller solution vector via interpolation. MATLAB supports linear interploation in the griddata() function, but I need to use splines.

I just stumbled across a similar question here: http://www.mathworks.com/matlabcentral/answers/76465-interpolation-of-4d-data