Generating and Evaluating City Building Facades Using Artificial Intelligence

Abstract

The present study examines the various methods involved in the generation of city building facades. Elements of the entire process are explored – from obtaining the necessary information to using it to create a harmonious urban environment. The methods considered – the recognition and generation of facade elements, the generation of as-built Building Information Model (BIM) and Digital Twins of buildings (DT) – are evaluated against the needs of the generation, as well as what their place is in the foreseeable future development of architecture in the metaverse. The purpose of the study is to present the best practices in the field depending on the needs of the model and to show their practical application. An example is given of using drone imagery to assess the predominant coloration of facades with a view to generating examples harmonious with surrounding buildings. The example uses the scripting language python and different libraries to evaluate the image information and presents the results as a figure displaying the three predominant colors of the given image.