Occupancy-informed: Introducing a method or flexible behavioural mapping in architecture using machine vision

The feedback-loops of the modern architectural design practice are broken, as architects rarely get to experience and learn about life in their buildings in a systematic way. Through novel survey and annotation methods, this project aims to develop tools and methodologies that assist architects in getting insights into the built environment. This paper describes the initial development of a framework for surveying and annotating occupant behaviour within architecture, called "Behavioural Situations". Using object recognition on embedded devices, it is possible to build an understanding of occupant behaviour, by coupling behavioural signifiers and their relations as nodes and edges in a graph representation.