What Urban Cameras Reveal About the City: The Work of the Senseable City Lab

Abstract

AbstractCameras are part of the urban landscape and a testimony to our social interactions with city. Deployed on buildings and street lights as surveillance tools, carried by billions of people daily, or as an assistive technology in vehicles, we rely on this abundance of images to interact with the city. Making sense of such large visual datasets is the key to understanding and managing contemporary cities. In this chapter, we focus on techniques such as computer vision and machine learning to understand different aspects of the city. Here, we discuss how these visual data can help us to measure legibility of space, quantify different aspects of urban life, and design responsive environments. The chapter is based on the work of the Senseable City Lab, including the use of Google Street View images to measure green canopy in urban areas, the use of thermal images to actively measure heat leaks in buildings, and the use of computer vision and machine learning techniques to analyze urban imagery in order to understand how people move in and use public spaces.