Polygonal Building Segmentation by Frame Field Learning
This addresses the gap between deep learning outputs and geographic information system requirements for vector data, though it is incremental as it builds on existing segmentation models.
The paper tackles the problem of converting raster image segmentations into vector polygons for building extraction from remote sensing images by adding a frame field output to a deep segmentation model, resulting in improved segmentation quality and facilitating polygonization.
While state of the art image segmentation models typically output segmentations in raster format, applications in geographic information systems often require vector polygons. To help bridge the gap between deep network output and the format used in downstream tasks, we add a frame field output to a deep segmentation model for extracting buildings from remote sensing images. We train a deep neural network that aligns a predicted frame field to ground truth contours. This additional objective improves segmentation quality by leveraging multi-task learning and provides structural information that later facilitates polygonization; we also introduce a polygonization algorithm that utilizes the frame field along with the raster segmentation. Our code is available at https://github.com/Lydorn/Polygonization-by-Frame-Field-Learning.