Shape Interpolation from csv

I have a shapes in terms of cartesian coordinates read from a csv file and want to interpolate the shape to increase the number of points I have. CSV has three columns (X, Y, Z) and each row is on point. All the documentation I have found (meshgrid, datagrid, ect) works by using vectors to create a grid of points, I already have the points but matlab will not let me manipulate the data while in 'double' format. How can I read in the data and interpolate the points to increae the density of my meshes?

%Read in file data

a=csvread('r1.csv');

%convert m to mm

X=a(:,1)/1000;

Y=a(:,2)/1000;

Z=a(:,3)/1000;

Then I want to interpolate between data points to increase the number of points I have.

Any help will be greatly appreciated.

Answer 1

[Walter Roberson]

File Exchange, John D'Errico, "interparc"

[Sam Hill]

a=csvread('r1.csv')
a1=a(:,1)/1000;
a2=a(:,2)/1000;
a3=a(:,3)/1000;
ra=interparc(0.5, a1, a2, a3)

returns the error:

Error using chckxy (line 50)

The first input must contain unique values.

Error in spline (line 72)

[x,y,sizey,endslopes] = chckxy(x,y);

Error in interparc (line 316)

spl{i} = spline(cumarc,pxy(:,i));

Answer 2

[Stephan]

If the curvature is small enough compared to the distance between the original points, a little bit of analytical geometry might help:

% Some data to play with - an arbitrary line in 3D
x = [1 2 3 4];
y = [-1 5 2 7];
z = [0 1 2 6];

% Concatenate vectors of x, y, z to one matrix
A = [x; y; z];

% Calculate Coordinates of the bisector to get query points
B = (A(:,1:end-1) + A(:,2:end)) ./ 2;

% Extract the single vectors
xq = B(1,:);
yq = B(2,:);
zq = B(3,:);

% plot the line that connects all original points
plot3(x,y,z)
hold on

% plot the original points in blue
scatter3(x,y,z,'ob','filled')

% plot the new interpolated points in red
scatter3(xq,yq,zq,'or','filled')
hold off

You could repeat this process (for example by using a function) if you need more points.