Spatial Video is any video sequence that has spatial properties associated with it. This paper presents a conceptual model called a Viewpoint which is used to fully define a Spatial Video data set. We highlight how existing models of Spatial Video are limited by their disparate requirements and how our Viewpoint definition can be used in the general case. To do this we define a model that uses the GIS point and polygon data type primitives to create a hypothetically maximum spatial extent for each image contained within a Spatial Video sequence. This concept is shown to both restrict and extend these GIS primitives, but also provide the means to broaden Spatial Video’s exposure to a larger number of GIS geospatial analysis operations. Thus, with our conceptual model of Spatial Video as a Viewpoint moving object we can further develop our theories into practical data modelling examples and natural or formal language domains.