Add Segment type in Model that will enable to point to a particular SpatioTemporal fragment of a Video MediaUnit
Localising an object on a particular frame of a Video in Model 1.1.x and 1.2 is time consumming. You need first to define a TemporalSegment on your Video. Create an Image MediaUnit to represent the extracted frame. Then you can localise the object on this frame through a SpatialSegment.
If we define a new SpatioTemporalSegment that have both spatial and temporal components it will ease this process.