Utilizing the Transformer Architecture Combined with EEGNet to Achieve Real-Time Manipulation of EEG in the Metaverse.

In the modern era, research on Brain-Computer Interfaces (BCIs) is thriving, with increasing focus on the application of Motor Imagery (MI). This study has developed a game aimed at collecting MI signals from subjects imagining movements of their left hand, right hand, and being in a rest state, capturing the Electroencephalography (EEG) signals 1 second before each imagined movement. To achieve real-time interpretation of these signals, this research employed the Transformer architecture for processing EEG signals. Thanks to its self-attention mechanism's strong adaptability to sequential data, the Transformer not only effectively eliminated noise but also successfully extracted key features. Subsequently, these Transformer-processed signals were inputted into EEGNet for a three-class classification. In tests conducted with eight participants, our model achieved an average accuracy of 71.32%, marking a 9.17% improvement over methods that only used EEGNet. This outcome not only demonstrates the potential of combining Transformer and EEGNet in BCI applications but also offers a new direction for future research.