Psychtoolbox>PsychKinectCore.{mex*} subfunction

[imageOrPtr, width, height, channels, extType, extFormat] = PsychKinect(‘GetImage’, kinectPtr [, imtype=0][, returnTexturePtr=0]);

Return the color image data for the frame fetched via ‘GrabFrame’.

If ‘returnTexturePtr’ is zero (default), a uint8 3D image matrix is returned.
The matrix is usually in RGB pixel interleaved row-major format for efficiency
reasons, ie. to improve processing speed for realtime applications.
If the optional ‘bayerFilterMode’ flag in PsychKinect(‘Open’) was set to 0, then
the image is returned as a row-major 2D matrix with the raw color sensor data
instead of post-processed RGB bayer filtered data, again for efficiency reasons.
A ‘bayerfilter’ setting of 2 would return a row-major 2D matrix with infrared
depth sensor raw data.
If you want to use the image matrix with standard Matlab/Octave or PTB
functions, e.g., feed it to the image processing toolbox, imwrite(), imshow(),
or Screen(‘MakeTexture’), you need to convert it into the expected Matlab/Octave
planar column-major format:

for i = 1:channels
retpixels(:,:,i) = transpose(squeeze(imageOrPtr(i, :, :)));
end
imshow(retpixels);

‘imtype’ if set to 1, returns a color-coded depth image instead of the RGB
camera image.
If ‘returnTexturePtr’ is one, a double-encoded memory pointer to a LUMINANCE8 or
RGB8 rectangle texture buffer is returned, for use with
Screen(‘SetOpenGLTextureFromMemPointer’) for injection into PTB’s texturing
system. ‘extType’ and ‘extFormat’ encode the proper parameters to pass for
external format and type to Screen(‘SetOpenGLTextureFromMemPointer’).
See for example KinectDemo.m for how this is used for fast video display.

###See also: