An image processing apparatus includes a memory that stores instructions, and a processor in communication with the memory. The processor executes the instructions to set a plurality of image sets according to a predetermined rule using a plurality of images obtained by continuous photography, to generate a first conversion coefficient corresponding to a moving amount of an object between images in each of the plurality of image sets, and to generate a second conversion coefficient used for an alignment process of a base image and a correction target image other than the base image by using a plurality of first conversion coefficients. The base image and the correction target image are included in the plurality of images.