A mobile entity tracking device is provided with: a video receiving unit for receiving a moving-image frame of a ball game captured by each of a plurality of cameras present at different positions; a mobile entity candidate extraction unit for extracting a candidate for a mobile entity using a plurality of moving-image frames; a mobile entity selection unit for displaying candidates for a mobile entity and accepting selection, by a user, of the mobile entity to be tracked; and a mobile entity tracking unit for tracking the mobile entity that is the object to be tracked. When the mobile entity selection unit accepts selection, by a user, of the mobile entity to be tracked, the mobile entity tracking unit corrects the object to be tracked to the mobile entity selected by the user.