lowvisionresearch / revas Goto Github PK

We need to keep track of the open children GUIs and close them when ReVAS is closed.

On several folders containing each on average 10 videos, there is each time 1-2 video not processed
for no apparent reason.

important, recurrent and unsolved

Hi,
it would be nice, as the code will be open source, that we use the same header like:

% ===========================================================
%  Function FilterStrips    by Mehmet Agaoglu and Matt Sit
%                Sight Enhancement Laboratory at Berkeley 
%                School of Optometry, University of California, Berkeley
%  Please cite this paper: 
%  doi:
% ===========================================================

Occurs when ReVAS is used as a standalone app (on macOS Mojave). When all pre-processing steps are disabled (for a previously analyzed video) and only the strip analysis is selected, the program errors out because it cannot find the reference frame file (*_dwt_nostim_gamscaled_bandfilt_refframe.mat) although it does exist in the folder where the video resides.

I found this page as potential reasons why this might be happening http://matlab.wikia.com/wiki/FAQ#Why_can.27t_my_standalone_compiled_executable_find_my_files.3F

After strip analysis at the Fine Ref stage, peak cross-correlations for each individual strip is plotted. But the horizontal axis is wrong. e.g., For a 1000Hz sampling rate in a 1 second video, the axis should be from 0 to 1 second but it goes from 0 to 1000 seconds.

I tried to run the GUI using the demo setup, and this failed with:

findblinkframes cannot come after stripanalysis.
06-Nov-2023 15:34:29: ERROR: Pipeline is invalid for _calib. Moving onto next file.

I do not have a stripanalysis item before findblinkframes, findblinkframes is the first thing in the pipeline. Stepping through the code, it turns out the problem is that the path to ReVAS (or the video i am processing, dunno which) contains the folder retinal_strip_analysis, which somehow triggers this first element to be added. I commented out some code and things are ok now, but this seems kinda crazy?

Hi,
I think it would be informative to add the unit in which are expressed the eyePositionTraces in the tutorial.
Because rereferencing is possible, I am inferring that unit are pixels, however because filtering options (saccades detection parameters) are expressed in visual degree, I am not sure if the eyePositionTraces are minutes of arc. Minutes are close to pixel here: for a 10 deg FoV, there are 488 pixels vs 600 minutes.