Rui-Zhe Guan

Global and Neighbor-Aware Token Learning for Weakly Supervised Remote Sensing Image Semantic Segmentation

Image-level weakly supervised remote sensing semantic segmentation aims to learn pixel-level land-cover prediction using only image-level labels, greatly reducing the annotation cost of fully supervised methods. Class activation map (CAM)-based methods are widely used for this task, but they usually focus on the most d...