search

prajwalsingh / EEGStyleGAN-ADA

Pytorch code of paper "Learning Robust Deep Visual Representations from EEG Brain Recordings". [WACV 2024]
MIT License
27 stars 4 forks source link

ThoughtViz and Object dataset #16

Closed jontyharrison closed 2 months ago

jontyharrison commented 3 months ago

Hi there, great paper!

I am trying to recreate the metric results on ThoughtViz EEG2Image but having issues with the data not being structured in the same way. I managed to produce the results on CVPR40.

In the link given in EEGCVPR40 Dataset Availability · Issue #1 · prajwalsingh/EEGStyleGAN-ADA (github.com) the data is not structured how it is in EEGStyleGAN-ADA/EEGStyleGAN-ADA_ThoughtViz/cmd_metricalc.txt at main · prajwalsingh/EEGStyleGAN-ADA (github.com) .

The code requires the image and EEG data to be linked but in that dataset it is not, I was wondering if there is a link to the data structured to match the .txt or a link to code to create that data structure.

Also in EEG2Feat for the Object dataset, the dataset in the link does not match what the code is expecting in https://github.com/prajwalsingh/EEGStyleGAN-ADA/blob/main/EEG2Feat/Triplet_LSTM/Object/config.py. Is the code to process the data available?

Many thanks!

prajwalsingh commented 3 months ago

Hi @jontyharrison , thank you for reading our work.

For the Thoughtviz dataset, kindly check this issue [link] and for the Object dataset, we have uploaded the split of the data that we have used in our work [Object Dataset Split] (code I am not able to find that we use to create the split, might have missed it while taking the project backup).