Convolutional Neural Networks Approach Comparison in EEG-Based BCIs
摘要
In this work, we explore and compare Deep Learning (DL) classifiers to the EEG signal acquired based on motor imagery paradigms. Previous papers have demonstrated the effective-ness of those neural networks on Brain-Computer interfaces studies. So, we compared well-known pre-trained architectures like VGG16 and built-up non-trained Convolutional Neural network. Finally, we discuss and propose perspectives and guides for the follow up of the work.