A MACHINE LEARNING APPROACH TO ESTIMATE WINDOWS-TO-WALL RATIO USING DRONE IMAGER

A building’s window-to-wall ratio (WWR) has critical influence on heat loss, solar gain, and daylighting levels, with implications for visual and thermal comfort as well as energy performance. However, in contrast to characteristics such as floor area, existing building WWRs are rarely available. In this work we present a machine learning based approach to parse windows from drone images and estimate the WWR. Our approach is based on firstly extracting the building 3D geometry from drone images, secondly performing semantic segmentation to detect windows and finally computing the WWR. Experiments show that our approach is effective in estimating WWR from drone images.