Machine-learned Regularization and Polygonization of Building Segmentation Masks
This work addresses the need for automated and aesthetically improved building polygon extraction in geospatial applications, representing an incremental advancement in segmentation post-processing.
The authors tackled the problem of generating realistic and rectilinear building outlines from segmentation masks by using a GAN for boundary regularization and a CNN for polygonization, achieving accurate and visually pleasing results across three datasets.
We propose a machine learning based approach for automatic regularization and polygonization of building segmentation masks. Taking an image as input, we first predict building segmentation maps exploiting generic fully convolutional network (FCN). A generative adversarial network (GAN) is then involved to perform a regularization of building boundaries to make them more realistic, i.e., having more rectilinear outlines which construct right angles if required. This is achieved through the interplay between the discriminator which gives a probability of input image being true and generator that learns from discriminator's response to create more realistic images. Finally, we train the backbone convolutional neural network (CNN) which is adapted to predict sparse outcomes corresponding to building corners out of regularized building segmentation results. Experiments on three building segmentation datasets demonstrate that the proposed method is not only capable of obtaining accurate results, but also of producing visually pleasing building outlines parameterized as polygons.