To generate a plurality of kinds of moving pictures on the basis of the result of feature value extraction of an object in a picture, for example, to compose a portrait picture to output portrait moving pictures expressing an expression change or the like.
The feature value of the picture is extracted by a feature value extraction means 101, and proper parts data corresponding to the feature value is extracted from a moving picture face parts data storage means 103 by a moving picture face parts data extraction means 102. Extracted parts data are arranged in positions determined for each face contour parts classification by a parts arrangement means 108 to compose a portrait picture. A user uses a moving picture display designation input means 111 to indicate the start of display of a moving picture presenting expression change or the like, continuous display of expressions, designation of parts to be displayed in a moving picture, the change of combination of moving picture data, or the like.
MATSUOKA ATSUO
KONYA MINEHIRO