An image processing apparatus comprises: a colored level calculation unit configured to calculate, for a first region in an image, an index indicating a strength of color; a feature amount calculation unit configured to calculate a first feature amount relating to luminance components in each of the first region and a second region, and a second feature amount relating to color components in the regions; an interest level calculation unit configured to calculate an interest level for the first region based on a difference for the first feature amount between the regions, a difference for the second feature amount between the regions, and an importance for each of the differences; and a region of interest detection unit; wherein the interest level calculation unit determines the importance for the differences based on the index.