Decoding visual brain representations from electroencephalography through Knowledge Distillation and latent diffusion models

TL;DR

This study introduces a novel EEG-based brain decoding method that combines knowledge distillation and latent diffusion models to classify and reconstruct images from neural activity, advancing visual cognition understanding.

Contribution

It presents a new approach integrating CNNs, CLIP-based knowledge distillation, and latent diffusion models for improved image classification and reconstruction from EEG data.

Findings

Achieved 80% top-5 accuracy in image classification from EEG

Enabled credible image reconstruction from neural signals

Outperformed standard CNN and RNN benchmarks

Abstract

Decoding visual representations from human brain activity has emerged as a thriving research domain, particularly in the context of brain-computer interfaces. Our study presents an innovative method that employs to classify and reconstruct images from the ImageNet dataset using electroencephalography (EEG) data from subjects that had viewed the images themselves (i.e. "brain decoding"). We analyzed EEG recordings from 6 participants, each exposed to 50 images spanning 40 unique semantic categories. These EEG readings were converted into spectrograms, which were then used to train a convolutional neural network (CNN), integrated with a knowledge distillation procedure based on a pre-trained Contrastive Language-Image Pre-Training (CLIP)-based image classification teacher network. This strategy allowed our model to attain a top-5 accuracy of 80%, significantly outperforming a standard CNN and various RNN-based benchmarks. Additionally, we incorporated an image reconstruction mechanism based on pre-trained latent diffusion models, which allowed us to generate an estimate of the images which had elicited EEG activity. Therefore, our architecture not only decodes images from neural activity but also offers a credible image reconstruction from EEG only, paving the way for e.g. swift, individualized feedback experiments. Our research represents a significant step forward in connecting neural signals with visual cognition.