search

JasperLS / Understanding_Videos_at_Scale

This repository includes the notebook of „Understanding Videos at Scale: How to Extract Insights for Business Research“
MIT License
21 stars 10 forks source link

Issue with variance measures #2

Open hdlucas opened 2 years ago

hdlucas commented 2 years ago

Hello, This looks like a fantastic tool. I am interested in quantifying variance for a set of video stimuli my lab is using in cognitive electrophysiology experiments. I tried out the notebook code on three of the videos, and while it worked for all of the other measures, the variance measures were not provided. The logs contained this message:

Found no valid file for the classes frames. Supported extensions are: .jpg, .jpeg, .png, .ppm, .bmp, .pgm, .tif, .tiff, .webp

Any help resolving this would be much appreciated. Thanks for your work on this!

DATE_logfile.csv

hdlucas commented 2 years ago

Hi again, I tried it again today and it worked for some reason! I don't know what was different. However, all of my variance measures were 1 because my videos only have one scene. Do you happen to know if the code can be easily adapted to capture within-scene variation?

casmocos commented 2 years ago

Hi, I am also using this tool to parse some videos. I also find that variance measures never work. I'm curious about that. Do you have any possible explanation for it? Thank you very much!