/looking-glass

Primary LanguageJupyter Notebook

looking-glass

Team: Catherine Rasgaitis, Srishti Bakshi

USAGE: 0. set subject name in app2 and app3 to set saving directory

  1. set INDEX_CHANNELS in app2 to select what nodes to record from
  • note that temporal resolution and number of nodes have a significant tradeoff
  1. run app2.py w/ eye tracker (show to subject) and run app3.py by running Petal Metrics first (this will run EEG collection in VSCode and does not open in browser)
  2. syncing step: subject should blink several times while facing away from the screen. We can use this data to sync the eye-tracking and eeg clocks
  3. app2.py will stop running automatically, you must press 'q' in terminal for app3.py to stop running. Data saves after both apps terminate.

NOTES:

  • app3.py to record eye-tracking and display images
  • app2.py to record EEG data

TODOs:

  • fix build_dataset function✅

  • add flagging functionality ✅

  • finish modifying StyleGAN with different anchor points and weighting parameters

  • conglomerate script -> input a json file of people objects with name, familiarity, image.

    • outputs a folder with correct images
  • finally input folder name with images to use in app ✅

  • build up UI for st app ✅

  • eye tracking ✅

  • eeg recording ✅

  • script should include info on initial calibration step with tobii