In the modern world, cities need to keep up with the demand for mobility, efficient infrastructure and environmental sustainability. The future smart cities use intelligent information and communication technologies to raise the quality of life. This includes computer vision as one of the main technologies. It can observe and analyse human activities from a distance in a noninvasive manner. Traditional computer vision utilises RGB cameras, but problems with this sensor include its light dependency, and the privacy issues that can be raised by people being observed. In this paper, we propose the use of thermal imaging in real-time smart city applications. Thermal cameras operate independently of light and measure the radiated infrared waves representing the temperature of the scene. In order to showcase the possibilities, we present five different applications which use thermal imaging only. These include both indoor
and outdoor scenarios with the purposes of people detection, counting and tracking, as well as one application for traffic safety evaluation.