In filter processing of a disparity map and a depth map, a disparity estimation accuracy is improved while maintaining the edge of a subject. An image processing apparatus that corrects a depth image representing information about a depth to a subject in a scene, which is the same scene of a plurality of images obtained by photographing the same subject from different viewpoints. The image processing apparatus includes: an evaluation value derivation unit configured to derive a matching evaluation value between the plurality of images; a weight determination unit configured to determine a weight in the correction based on the evaluation value; and a correction unit configured to correct a pixel value in the depth image by using the weight.