Certain embodiments of these teachings take inputs of a) an image to be segmented and b) a set of user input markers that identify at least a foreground of said image. From these inputs are determined strong foreground regions Sf, strong background regions Sb and ambiguous regions G. Each ambiguous region Gi is classified as either a weak foreground region Wf or a weak background region Wb by comparing color of each ambiguous region Gi against color of regions adjacent thereto and merging the respective ambiguous region Gi with its adjacent region having a most similar color. The strong foreground regions Sf are also merged with the weak foreground regions Wf. What is output therefrom is a foreground image of the image to be segmented, separate and distinct from any region that is classified as one of the strong background regions or weak background regions.