gkaguirre
commented
6 years ago This aspect of the code is under active development. There was some code developed by Giulia that handled the temporal support. I am re-writing this.
In the current / new approach, a timebase.mat file is generated at the stage of video deinterlacing. In principle, the timebase file could be from any other source. It is a .mat file that contains the structure timebase that then has the field values, which is a nx1 vector, with n being the number of frames in the video. The timebase should be specified in units of milliseconds. The value of zero could be assigned to the timebase at (e.g.) the start of the scan, or the stimulus presentation.
The analysis of the gazeCal files (and extracting targets) is yet another topic, and something that is only just working now. Might be worth a Skype session to chat about how to work with these.
Hi all,
I am attempting to use this toolbox to apply gaze and size calibrations to out eyetracking data from transparentTrack but am running into an issue (or a few). I am attempting to derive a timebase for the GazeCal run so that I can use the selectDotTimesGUI function in transparentTrack. However, in line 129 of timebaseWrapper, called from SCRIPT_deriveTimeBase, it throws an error saying the function deriveTimebaseFromLTData does not exist. Should this file be included or is the name of the function changed?
I apologize for making this long and including another question, but I was also curious what the main control script for calculating gaze and size calibrations and applying them to runs already analyzed via transparentTrack would be. The SCRIPT files seem similar in their their scope from a naive perspective and I haven't been able to figure out the best way to adapt the toolbox to our setup yet. Any advice would be appreciated. Thank you!