Our processed LoveDA dataset is used for the paper "LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images"
These trained deep learning are trained to extract building footprints from high-resolution aerial imagery and LiDAR-derived nDSM with a spatial resolution of 20cm or less. These models were used to e
We randomly selected more than 2,100 aerial images with a size of 400×416 pixels from the AIRS dataset as our image dataset, And among them, 100 images were used for testing. To create annotation file
