Hi Wakefield,
the python code for extracting the uploaded images works well for the .csv files created by pavlovia surveys (where responses from different partciapnts are all collated into a single file). However, to deal with the display bug on android phones I have now had to embed the picture upload survey as a survey component into a psychpy experiment as shown here (Pavlovia Surveys display not correct on mobile devices after update - #8 by wakecarter)
My problem is that your script does not work for the .csv files I download from pavlovia. On pavlovia, you get a .csv file for each particiapnt instead of a single .csv for all participants., and the file structure seems to be different. Is there a way to get the equivalent of the single .csv generated by pavlovia surveys when said survey is embedded?
Best wishes,
Henning