To more naturally and sharply process a photographed image.

A motion vector calculation unit 12 extracts a motion vector for the unit of each pixel from each of a plurality of input images, an occlusion processing unit 14 generates a direct motion vector for a pixel on an image to be processed for each pixel in the plurality of input images based on the extracted motion vector, and an image composing unit 15 composes the plurality of input images based on the direct motion vector for pixel values of pixels in the plurality of input images. The present invention may be applicable to video cameras.