Iterative Deep Learning for Road Topology Extraction
This work solves the problem of automated road mapping for urban planning and medical imaging, though it is incremental as it builds on existing dense classification models.
The paper addresses road network topology extraction from aerial images by using a CNN to predict local connectivity and iteratively infer global topology, achieving strong performance on road networks and generalizing to retinal vessel networks.
This paper tackles the task of estimating the topology of road networks from aerial images. Building on top of a global model that performs a dense semantical classification of the pixels of the image, we design a Convolutional Neural Network (CNN) that predicts the local connectivity among the central pixel of an input patch and its border points. By iterating this local connectivity we sweep the whole image and infer the global topology of the road network, inspired by a human delineating a complex network with the tip of their finger. We perform an extensive and comprehensive qualitative and quantitative evaluation on the road network estimation task, and show that our method also generalizes well when moving to networks of retinal vessels.