Motion information for an object and background information are broadcast-transmitted on a transmitting side. On a receiving side, the motion information and the background information are received, and one of a plurality of character models previously stored is selected. A character image is generated using the selected mode and the received motion information. The character image and the received background image are synthesized and displayed on one screen. The viewer can select a character of his or her own liking, and can view the broadcast-transmitted program as a program in which the particular character appears.