According to zoologist Andrew Parker, about 550 million years ago, there was an explosion of life forms on the planet, all the major animal groups rapidly and dramatically appeared. Parker’s controversial but increasingly accepted ‘Light Switch Theory’ states that it was the development of vision in primitive animals that caused the biological explosion. Our built environment has more visual data than ever before, and cameras will continue to be installed within the built environment. All these visual data were previously lost until advancements in computer vision and compute power allowed for accurate neural networks to be developed with consumer computing. The built environment has an opportunity to use computer vision to optimise building energy performance by understanding building usage with real-time occupant data and computer vision applications. This research project entailed the development of a computer vision application that detects and counts occupants within live video feeds. This was achieved by training a convoluted neural network on a dataset of 15,000 images of ‘human bodies’ extracted from Googles Open Images v6. The computer vision application was used to count the number of occupants in lecture venues by leveraging the existing venues’ CCTV camera. A medical grade air quality device was placed within the assessed lecture venue and real-time occupant count was correlated against real-time indoor air quality data. The aim of this research was to determine how computer vision can be used to generate accurate real-time occupant data to improve the management of facilities and the indoor building environment. The results indicate that computer vision and the ability to draw data insights from image and video sources can significantly improve the management of facilities by providing the facility manager real-time data and an opportunity to improve facility management tasks through automation.