retval = SetAnaglyphStereoParameters(subcommand, windowPtr, [red green blue]);
Change parameters of built-in anaglyph stereo display function at
runtime. Allows to change all relevant settings of the algorithm at any
time in your script. You need to open your onscreen window with the
‘stereomode’ parameter set to one of the anaglyph modes (i.e. one of
6=Red-Green anaglyph, 7=Green-Red, 8=Red-Blue, 9=Blue-Red, where
Red-Green means “Left eye displayed in red channel, right eye displayed
in green channel”).
Anaglyph stereo also works in stereo modes 2, 3, 4, 5 and 10 if you used
PsychImaging(‘AddTask’, ‘LeftView’, ‘DisplayColorCorrection’, ‘AnaglyphStereo’)
… and …
PsychImaging(‘AddTask’, ‘RightView’, ‘DisplayColorCorrection’, ‘AnaglyphStereo’)
… setup sequence to add an anaglyph shader to the end of each view
channel. However, in those “non-pure” exotic anaglyph modes only a subset
of the commands mentioned here work, specifically the ones for advanced
anaglyph display, ie., subfunctions ‘ColorAnaglyphMode’ and friends. Inverted mode
or simple mode or changing NTSC weights or simple gains does not work in
Additionally you also need to enable the Psychtoolbox image processing
pipeline by using the PsychImaging() command for configuring and opening
See ‘help PsychImaging’ and ‘help PsychGLImageProcessing’ for an overview
of the imaging pipeline.
Parameters and their meaning
‘windowPtr’ is the handle to a anaglyph stereo onscreen window.
‘subcommand’ can be one of:
‘LeftGains’ - Provide (in optional vector [red green blue]) per-channel
color gains for computation of the left-eye channel. This allows to
adjust the output colors to optimally match the color emission / filter
profile of your anaglyph glasses and monitor - Reduce ghosting / visual
SetAnaglyphStereoParameters(‘LeftGains’, win, [1.0 0.0 0.0]); sets the
red-gain for the left eye image to 1.0 and disables output of the left
eye image into the green or blue channels. This is the default setting
for Red-Green or Red-Blue anaglyph, left eye view only goes into red
The command always returns the old gains as a 3 component vector.
‘RightGains’ - See left gains, but this time for right image:
SetAnaglyphStereoParameters(‘RightGains’, win, [0.0 0.7 0.0]);
would set the green channel to 70% output intensity and disable the red-
and blue channels. This is a typical setting for Red-Green anaglyph
stereo: Right eye only written to green color channel, but with reduced
output intensity to compensate for the stronger green-sensitivity of the
‘ColorToLuminanceWeights’ - Set the color to luminance conversion
weights: PTB converts all color images into pure luminance (greyscale)
images before distributing them into the color channels via the formula:
Luminance = redvalue * redweight + bluevalue * blueweight + greenvalue *
The default red- green- and blue-weights are (0.3, 0.59, 0.11), so:
L = 0.3*red + 0.59*green + 0.11*blue. This is according to NTSC standard
(if i remember correctly!). You can change the weights via:
SetAnaglyphStereoParameters(‘ColorToLuminanceWeights’, win, [redweight greenweight blueweight]);
The command always returns the old weights as a 3 component vector.
‘BackgroundColorBias’ - Set the “background color” for inverted anaglyph
mode. This color value is added to each pixel. It defaults to zero - bias
free display. For inverted mode, one usually sets it to maximum output
intensity, so that the default background is a bright output.
‘InvertedMode’ - Switch to inverted display mode, i.e., set the
‘BackgroundColorBias’ and color gains properly. The image is displayed in
an inverted fashion, like a photo-negative. This allows for less ghosting
artifacts if your stimulus is sparse, e.g., dots and lines. This trick
was proposed by Massimiliano di Luca.
‘NormalMode’ - Switch from inverted mode to normal anaglyph mode.
Background color bias is set to zero, gains a positive.
Color reproduction in color images with Anaglyphs:
This switches to a more flexible anaglyph implementation, where the
output color is calculated as:
where [redout, greenout, blueout] == Final anaglyph color pixel.
[leftred, leftgreen, leftblue] == Input left eye pixel color.
[rightred, rightgreen, rightblue] == Input right eye pixel color.
If ‘Gamma’ is provided, then the ‘redout’ color is gamma corrected with
the given ‘Gamma’ value, ie. redout = redout ^ Gamma.
Multiple sets of possible matrices and gamma weights are already defined
and selectable with the following names:
The idea and matrices/gammas for these four selectable presets are taken
from the following website:
If you want to have an additional monoscopic overlay window, e.g., for
text instructions to the subject, fixation dots etc., you can get one via
a call to:
overlaywindow = SetAnaglyphStereoParameters(‘CreateMonoOverlay’, win);
This will create a fullscreen ‘overlaywindow’. You can draw to that
window like to any other window. It’s content will be overlaid to the
final anaglyph image, but without applying anaglyph conversion to it.
Wherever this overlay has an alpha color value of zero, the stimulus will
be shown. Wherever its alpha value is greater than zero, the overlays
image will be shown.
overlaywindow = SetAnaglyphStereoParameters(‘CreateGreenOverlay’, win);
… and …
overlaywindow = SetAnaglyphStereoParameters(‘CreateBlueOverlay’, win);
… will allow to only place the overlay info into the green color channel
or blue color channel, ie., the channel not used by Red-Blue or Red-Green
color anaglyph stereo.
‘GetHandle’ - Return GLSL handle to anaglyph shader. Allows to modify the
shader itself, e.g., replace it by your own customized shader. Only for