Interpolate matrix in 4D

I have a matrix data of size 20,10,20,10.

It is a 3D object that changes across time (10 time points which represent 10 seconds)

I would like to create a smooth transition interpolating values to subdivide those 10 seconds into e.g., 10th of a second, so 100 time points in total.

How to go about this in Matlab?

Answer 1

[Matt J]

You can use griddedInterpolant (preferable IMHO) or interpn.

[Matt J]

You can also use imresizen from this FEX download,

https://www.mathworks.com/matlabcentral/fileexchange/64516-imresizen-resize-an-n-dimensional-array?s_tid=srchtitle

A=rand(20,10,20,10);
Ainterp=imresizen(A,[1,1,1,10]);

whos A Ainterp

  Name          Size                     Bytes  Class     Attributes

  A            20x10x20x10              320000  double              
  Ainterp      20x10x20x100            3200000  double              

This doesn't subdivide the sample locations, though. It increases the fineness of the sampling without necessarily preserving the original samples.

[Tahariet Sharon]

Thanks. I don't understand this function. what does the [1,1,1,10] mean?

I want to preserve the values in A(:,:,:,1) and in A(:,:,:,10), which must be identical to the values in the Ainterp(:,:,:,1) and Ainterp(:,:,:,100). And the new values should be interpolated accordingly.

[Matt J]

what does the [1,1,1,10] mean?

It means make the 4th dimension 10 times bigger.

I want to preserve the values in A(:,:,:,1) and in A(:,:,:,10), which must be identical to the values in the Ainterp(:,:,:,1) and Ainterp(:,:,:,100). And the new values should be interpolated accordingly.

griddedInterpolant will do that.

A=rand(20,10,20,10);
F=griddedInterpolant(A);

Ainterp=F({1:20,1:10,1:20,linspace(1,10,100)});

whos A Ainterp

  Name          Size                     Bytes  Class     Attributes

  A            20x10x20x10              320000  double              
  Ainterp      20x10x20x100            3200000  double