3D Volume Interpolation with ba_interp3: Fast interp3 replacement

Faster Replacement for interp3

Description

ba_interp3 is a fast interpolation routine for 3D volumes. It can be used as a drop in replacement for matlabs interp3 function. It supports nearest neighbour, tri-linear, and tri-cubic interpolation.

Restrictions: The only restriction is, that the data has to be on a regularly spaced grid (i.e. coming from meshgrid.

Enhancements: The first advantage of ba_interp3 is that it is a lot faster.
ba_interp3 is up to 10 times faster than matlabs interp3 in realistic scenarios

Additionally, it can interpolate layered data. If you interpolate a matrix of size MxNxO x L1xL2x...xLk then each of the L1*L2*...*Lk layers is interpolated simultaneously. Think of this as in RGB images, where there would be a single additional dimension of size three.
An volume of 3x3 Matrices (i.e. the volume derivative) would have two additional dimensions of size three.

Acknowledgements

The author wishes to acknowledge the following in the creation of this submission:
Image Interpolation (ba_interp2)

• MATLAB release: MATLAB 7.6 (R2008a)
• Other requirements: mex compiler

[Patrik]

Great job, but I cannot compile!
The error is:

lcc preprocessor error: C:\ba_interp3\ba_interp3.cpp:62 Could not find include file <iostream>

What's the problem? Is it possible for you to include a compiled version for 32bit windows? Any help appreciated!

[Nicholas]

In order to compile for Windows using VS 2005 Professional and the C90 library, I had to change 'round(x)' to 'int(x+0.5)' and modify the string comparison terms as my compiler couldn't handle the 'string1 == string2' statements. Now it works well.

[vick Audi]

Hi Nicholas, could you please let me know what are those changes you made. I'm getting exactly the same error.

I'm a novice in C++ . I got rid of the round error by search-replace "round(" to "int(" but don't know how to fix the == error.

Thanks in advance for your input

[Hemal Naik]

Linear interpolation is faster and cubic is much faster, However the slice shown has more data then what we usually get by using matlab function Interp3.
For some applications you might want to have same looking slice but much faster, I customized it for my application, cubic is almost 60%-70% faster then intep3 with mri data set of matlab.
Very cool contribution, thanks for uploading this :-)

[David Johnson]

When I use mex -largeArrayDims to compile this function, the results are wrong (all zero). It works OK without this switch, though -largeArrayDims will probably become mandatory soon!

Matlab R2009a, Visual Studio 2008 SP1, Win7 x64 bit Enterprise.

Anyway, I really appreciate your work. Saved me a lot of time from coding this myself!