Highly accurate estimation results are obtained even though cameras used for shape estimation of an object are distributed in accordance with a plurality of points of interest. The image processing apparatus of the present invention includes: an estimation unit configured to estimate an object shape of an object within a multi-viewpoint video image captured by each of a plurality of camera groups in units of camera groups; and an integration unit configured to integrate estimation results of the object shapes estimated in units of camera groups based on a camera map indicating a position relationship between common image capturing areas in each of the plurality of camera groups.