Window detection in complex facades

Abstract

The complexity of a building façade provides challenge for window detection algorithms. Especially in the gradient projection approach, the presence of gradients outside window areas significantly reduces the quality of results. We present a modified gradient projection method robust enough to process complex façades of historical buildings. In a single image scenario, this method is able to provide results even for facades under severe perspective distortion. Our algorithm is able to detect many different window types and does not require a learning step. In this paper, we also extend this method into a multi-view scenario. We examine the results of the method in multi-view and evaluate its benefits.