Abstract:

The foreground group in a scene may be \'discovered\' and computed as a factorized approximation to the pairwise affinity of the elements in the scene. A pointwise approximation of the pairwise affinity information may in fact be interpreted as a \'saliency\' index, and the foreground of the scene may be obtained by thresholding it. An algorithm called \'affinity factorization\' is thus obtained which may be used for grouping. The affinity factorization algorithm is demonstrated on displays com- posed of points, of lines and of brightness values. Its relationship to the Shi-Malik normalized cuts algorithms is explored both analytically and experimentally. The affinity factorization algorithm is shown to be com- putationally efficient (O(n) floating-point operations for a scene com- posed of n elements) and to perform well on displays where the back- ground is unstructured. Generalizations to solve more complex problems are also discussed.