Abstract: The performance of a classifier in a brain-computer interface (BCI) system is highly dependent on the quality and quantity of training data. Typically, the training data are collected in a ...
Abstract: This paper introduces a simple yet powerful channel augmentation for visible-infrared re-identification. Most existing augmentation operations designed for single-modality visible images do ...