Journal Club
Stephen M
4/21/21
The Hope
Demonstrate the feasibility of decoding human thought by translating brain activity into an image
The Plan
- Train an end-to-end mental image generation framework
- General architecture is to use a VAE to map brain signals and images to a latent space, and then "align" these latent spaces
The Data
- Two sessions: image presentation session and a mental imagery session
Encoder Overview
Auto Encoder
Variational AE
Variational AE
\(\alpha\)-GAN
- Combines VAE with GAN
Combining The Approaches
Framework
EEG Encoder
fMRI Encoder
Image Reconstructor
Overall, a pretty simple idea. I don't think the architecture is too convoluted
Now that we have a "latent space" that is created from the image features, we can look at the other components
Encoders
- Literature on EEG signal extraction
Unlike the EEG signal which can be directly used to train the learning model [...]
Encoders use the same architectures
Encoders
We firstly employ Pearson Correlation Coefficient to select k most related features of the EEG signal or the fMRI signal according to each dimension of the m dimension image feature representation vector
Then, we construct m
parallel Bayesian regression sub-models. Each Bayesian regression predicts the
value of one corresponding dimension of the image feature representation vector
based on k most related features of the EEG signal or the fMRI signal
Data
- fMRI - EEG
- 5 subjects. Pretty low resolution imaging
- Visual Stimuli
- 20 categories with 50 examplars per category
- Data collection
- 5 sessions for image experiment
Training
- \( \alpha\)-GAN architecture is given in text
- Training doesn't make sense to me
- What atlas do they use?
- Bad assumptions
- Bad model
- Asses model using SSIM and MSE
Results
Fig 4
Summary and Discussion
- There is talk about split-test data on training?? No validation?
- What about the imagining dataset? Never talks about it
Thoughts
- I would NOT accept this paper
- Methods needs more detail
- Really coarse approach
I think that there is a better way:
HCP (along with Haxby) has publically available datasets of people watching movies
VAE-GAN Brain Decoder Paper
By smazurchuk
VAE-GAN Brain Decoder Paper
This is a slide deck I made for a journal club presentation. It presents an arxiv pre-print that tries to "decode human imagination"
- 62