A method of coding a sequence of images comprises at least one image comprising an image region of interest.The coding method comprises the following steps implemented for said at least one image of said sequence of images: determining (S10 to S23) an image portion comprising the image region of interest on the basis of a rate-distortion criterion representing a compromise between the compression efficiency the distortion of the image portion and the compression efficiency the distortion of the whole of said at least one image; and coding (S24) said at least one image on the basis of the determined image portion.