Transfer learning and SpecAugment applied to SSVEP based BCI classification

Open Webpage

Pedro R. A. S. Bassi, Willian Rampazzo, Romis Attux

Published: 01 Jan 2021, Last Modified: 14 May 2024Biomed. Signal Process. Control. 2021EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•SpecAugment can improve SSVEP based BCI classification performances.•DNNs surpass FBCCA and SVMs in a single-channel BCI with small data length.•Transfer learning made training faster.•Window slicing can be combined with SpecAugment to improve performances.