Perceptual Grouping Using Global Saliency -Enhancing Operatorst

We introduce saliency-enhancing operators capable of highlighting features which are considered perceptually relevant. We are able to extract salient curves and junctions ana' generate a description ranking these features by their likelihood of coming from the original scene. We suggest the global extenswnjleld as means of describing the behavior of a curve segment, in terms of its continuation. We show that a directional convolution of an edge image with the above field can produce useful descriptions. Other fields are also used in the same manner to produce similar results for domain-specifrc applications. The scheme is particularly usejid and robust as a gap filler and in the presence of noise. It is interesting to note that all operations are parameter-free, non-iterative and the processing is linear in the number of edges in the input image.