Guess What I Think: Streamlined EEG-to-Image Generation with Latent Diffusion Models
Paper • 2410.02780 • Published • 1
image imagewidth (px) 224 224 | conditioning_image array 2D | caption stringclasses 4
values | eeg_no_resample array 2D | label int32 0 3 | subject int32 0 23 |
|---|---|---|---|---|---|
[[0.7012389302253723,0.9251349568367004,0.996878445148468,0.9763063192367554,0.9341191053390503,0.91(...TRUNCATED) | image of a Apple | [[0.9937775731086731,0.994256317615509,0.9897092580795288,0.9791792035102844,0.9703242182731628,0.96(...TRUNCATED) | 0 | 0 | |
[[0.6836830377578735,0.905362606048584,0.9765087366104126,0.9563286304473877,0.9147791266441345,0.89(...TRUNCATED) | image of a Apple | [[0.9756802320480347,0.9797735810279846,0.9814591407775879,0.98531174659729,0.9889236092567444,0.990(...TRUNCATED) | 0 | 0 | |
[[0.6962940692901611,0.919508159160614,0.9909971356391907,0.9704274535179138,0.9282986521720886,0.91(...TRUNCATED) | image of a Apple | [[0.9724354147911072,0.9707574844360352,0.9635668396949768,0.9568552374839783,0.9630873799324036,0.9(...TRUNCATED) | 0 | 0 | |
[[0.6990741491317749,0.9227814674377441,0.9945699572563171,0.9741784930229187,0.932192325592041,0.91(...TRUNCATED) | image of a Apple | [[0.990363597869873,0.9910863637924194,0.9857866168022156,0.9939772486686707,0.990363597869873,0.988(...TRUNCATED) | 0 | 0 | |
[[0.686455488204956,0.9081469774246216,0.9788839817047119,0.9580240249633789,0.915701150894165,0.897(...TRUNCATED) | image of a Apple | [[0.971397876739502,0.9637706875801086,0.968299150466919,0.9702059626579285,0.9659157991409302,0.968(...TRUNCATED) | 0 | 0 | |
[[0.7004029154777527,0.9239842891693115,0.995394766330719,0.974479079246521,0.9319462180137634,0.913(...TRUNCATED) | image of a Apple | [[0.9848393797874451,0.9814702868461609,0.9906148314476013,0.9860425591468811,0.9802671074867249,0.9(...TRUNCATED) | 0 | 0 | |
[[0.6986591815948486,0.9222425818443298,0.9939178228378296,0.9734273552894592,0.9313586950302124,0.9(...TRUNCATED) | image of a Apple | [[0.9841631650924683,0.9853630661964417,0.9808039665222168,0.9736053347587585,0.9714457988739014,0.9(...TRUNCATED) | 0 | 1 | |
[[0.699134886264801,0.9225639700889587,0.9939346313476562,0.9730348587036133,0.9305081367492676,0.91(...TRUNCATED) | image of a Apple | [[0.9862717986106873,0.9788055419921875,0.9780829548835754,0.9857900142669678,0.9872351884841919,0.9(...TRUNCATED) | 0 | 0 | |
[[0.7041523456573486,0.9283137321472168,1.0,0.9791625738143921,0.9366379976272583,0.9182363152503967(...TRUNCATED) | image of a Apple | [[0.9966270327568054,0.9908446669578552,0.9896398186683655,0.994217574596405,0.9884352087974548,0.99(...TRUNCATED) | 0 | 1 | |
[[0.6949979066848755,0.9181689620018005,0.9897969365119934,0.9694889783859253,0.927672803401947,0.90(...TRUNCATED) | image of a Apple | [[0.9775093197822571,0.9830123782157898,0.9873191714286804,0.9851656556129456,0.9846872687339783,0.9(...TRUNCATED) | 0 | 0 |
This dataset is used in the paper "Guess What I Think: Streamlined EEG-to-Image Generation with Latent Diffusion Models" available on arxiv and the code is available at this repository.
The dataset is constructed to be used in controlnet scenarios.
Original paper of the dataset is here here
Original EEG data from here
Original images used to train the models from here