Various aspects of a system and method to generate a multi-dimensional video are disclosed herein. In accordance with an embodiment, the system includes an application server, which receives a plurality of videos feeds from a plurality of image-capturing devices that capture a pre-defined area. Further, sensor data is received from a plurality of sensors associated with one or more subjects in the pre-defined area. Thereafter, a location-of-occurrence of an upcoming activity-of-interest is predicted in the predefined area based on the received plurality of video feeds, sensor data, and/or pre-stored statistical data of the historical performance of the one or more subjects. One or more control instructions are communicated to one or more of the plurality of image-capturing devices to focus towards the predicted location-of-occurrence to enable the generation of the multi-dimensional video. Thereafter, one or more views of the generated multi-dimensional video are broadcasted to a plurality of subscriber-terminals.