作者: Florian Yger
DOI: 10.1109/MLSP.2013.6661972
关键词:
摘要: Recently, covariance matrices have been shown to be interesting features for signal classification and object detection. In this paper, we review compare the existing kernels on explore their use EEG in Brain-Computer Interfaces (BCI). This study addresses both experimental theoretical aspects of problem. Beside apparent complexity kernels, show that approach simplifies whole BCI system. Finally, empirically demonstrate simpler obtains state-of-the-art results competition IV dataset 2a.