1 paper
Nouf Almesafri, Hector Figueiredo, Miguel Arana-Catania
This study investigates the performance of the two most relevant computer vision deep learning architectures, Convolutional Neural Network and Vision Transformer, for event-based c…