MATLAB - Going from a disparity map to point cloud from stereo images on HELP!

Hi all, I need help with my stereoscopic 3-D reconstruction project, I'm currently stuck on the reconstructing a 3-D point cloud part. I have been given a pair of stereo images of the back of an eye, and I need to produce a 3D point cloud from them, I have got up to the disparity map, not sure if its optimal?

Here is what I've done so far:

[Hanieh Shabanian]

could you find any solution to your problem?

Answer 1

[Gautham Sholingar]

Another possible way to approach this problem is to use the following function from the Computer Vision System Toolbox which allows you to create a real-world 3D co-ordinate model from the disparity data:

https://www.mathworks.com/help/vision/ref/reconstructscene.html

As far as being optimal, you will need to explain what optimal means and evaluate this yourself. Try comparing the derived point-cloud to an expected model and see if this helps you evaluate the result.

[Haseeb Majid]

I have that toolbox installed, but how would I produce a point cloud from that disparity map?

[UBAID ULLAH]

I am running your code but i am getting a error. can you help me to solve it.

Error using zeros Requested 55033x54541 (11.2GB) array exceeds maximum array size preference. Creation of arrays greater than this limit may take a long time and cause MATLAB to become unresponsive. See array size limit or preference panel for more information.

Error in cvalgMatchFeatures>metricSSD (line 507) scores = zeros(N1, N2, outputClass);

Error in cvalgMatchFeatures>exhaustiveDistanceMetrics (line 186) scores = metricSSD(features1, features2, N1, N2, outputClass);

Error in cvalgMatchFeatures>findMatchesExhaustive (line 137) scores = exhaustiveDistanceMetrics(features1, features2, N1, N2, outputClass, metric);

Error in cvalgMatchFeatures (line 52) [indexPairs, matchMetric] = findMatchesExhaustive(features1,features2, ...

Error in matchFeatures (line 177) cvalgMatchFeatures(features1, features2, ...

Error in disparitymap (line 23) indexPairs = matchFeatures(features1, features2);

this was the error

Answer 2

[Munin]

@Hanieh Shabanian @Haseeb Majid the solution is simple:

disparityMap(disparityMap > 12) = 0;