B. Pan

FAViT: An Embedding Fusion of Vision Transformer and Attention U-Net for Landslides Segmentation Using Sentinel-1

This research introduces fusion attention vision transformer (FAViT), a novel hybrid deep learning framework for the automated rapid mapping of landslides from multitemporal Sentinel-1 synthetic aperture radar (SAR) amplitude data. A significant challenge in this domain is the need for methods that are both computation...