Depth Ordering and Figure-Ground Segregation in Monocular Images derived from Illusory Contour Perception
Marcus Hund, Bärbel Mertsching · European Conference on Artificial Intelligence · 2006
One of the elaborated tasks for the low level image processing stage of a vision system is the completion of cluttered or occluded object boundaries and the depth assignment of overlapped boundaries. We describe a novel method for depth ordering and figure-ground segregation from monocular depth cues, namely the arrangement of so-called illusory contours at junctions in the edge map of an image. Therefore, a computational approach to the perception of illusory contours, based on the tensor voting technique, is introduced. Furthermore, two simple rules for figure-ground segregation are presented that are capable of explaining a wide range of perceptual phenomena like the Kanizsa illusion. Finally, we present a diffusion approach to derive the depth ordering for the whole contour from the figure-ground segregation at key points.