An image processing apparatus obtains a difference between a partial region of an input image and a partial region of a background image, stored in a storage device, corresponding to the partial region of the input image, determines whether each partial region of the input image is a moving body region or a background region based on the difference, and combines the partial region of the input image determined as the background region and the partial region of the background image corresponding to the partial region of the input image to update the background image.