Dummy mode of GazeParser.TrackingTools for standalone testing

注釈

  • VisionEgg backend is no longer supported by 0.11.0 or later.

GazeParser.TrackingTools has dummy mode to help writing scripts when SimpleGazeTracker is not available. To use dummy mode, create an controller object with dummy=True option.

from GazeParser.TrackingTools import getController

#PsychoPy
tracker = getController(backend='PsychoPy',config='TrackerSettings.cfg', dummy=True)

#VisionEgg
tracker = getController(backend='VisionEgg', config='TrackerSettings.cfg', dummy=True)

The dummy controller shows black filled rectangle when camera preview and calibration results are requested (Figure 1 and 2).

../_images/dummy001.png

Figure 1

../_images/dummy002.png

Figure 2

The dummy controller output commands to the standard output instead of sending commands to SimpleGazeTracker. This is an example of output:

connect (dummy)
openDataFile (dummy): participant001.csv
sendSettings (dummy)
Dummy sendCommand: toggleCalResult
Dummy sendCommand: startCal -409,-307,409,307
Dummy sendCommand: getCalSample 307,230
Dummy sendCommand: getCalSample 0,230
Dummy sendCommand: getCalSample -307,0
Dummy sendCommand: getCalSample -307,230
Dummy sendCommand: getCalSample 307,-230
Dummy sendCommand: getCalSample -307,-230
Dummy sendCommand: getCalSample 0,0
Dummy sendCommand: getCalSample 0,-230
Dummy sendCommand: getCalSample 307,0
Dummy sendCommand: endCal
startMeasurement (dummy):
stopMeasurement (dummy):
startRecording (dummy): trial1
sendMessage (dummy): STIM 0 0
sendMessage (dummy): STIM -300 -200
sendMessage (dummy): press space
sendMessage (dummy): STIM 200 -200

(snip)

sendMessage (dummy): STIM 100 100
sendMessage (dummy): STIM -300 200
sendMessage (dummy): STIM 100 -100
stopRecording (dummy): end trial
close (dummy)

The dummpy contoller's getEyePosition() returns mouse cursor position instead of current gaze position. Gaze-contingent stimulus presentation can be tested using a mouse.

注釈

Currently, units of the windows must be 'pix' or the same to that of getEyePosition() when PsychoPy backend is used (VisionEgg backend does not support 'units').