Perceptual Organization of Occluding Contours
of Opaque Surfaces
CVIU Special Issue on Perceptual Organization in Computer Vision
Eric Saund
Xerox Palo Alto Research Center
Abstract
This paper offers computational theory and an algorithmic
framework for perceptual organization of contours bounding opaque
occluding surfaces of constant lightness. For any given visual scene, a
sparse graph is constructed whose nodes are salient visual events such
as contrast edges, and L-type and T-type junctions of contrast edges,
and whose arcs are coincidence and geometric configurational relations
among node elements. An interpretation of the scene consists of choices
among a small set of labels for graph elements reflecting physical
events such as corners, visible surface occlusion, amodal continuation,
and surface occlusion sans visible contrast edge (which perceptually
give rise to illusory contours). Any given labeling induces an energy,
or cost, associated with physical consistency and figural interpretation
biases. Using the technique of deterministic annealing, optimization is
performed such that local cues propagate smoothly to give rise to a
global solution. We demonstrate that this approach leads to correct
interpretations (in the sense of agreeing with human percepts) of
popular simple ``Colorforms'' figures known to induce illusory contours,
as well as more difficult figures where interpretations acknowledging
accidental alignment are preferred.
paper: pdf
manuscript: pdf
Back to Eric Saund's Papers Page.
Back to Eric Saund's Home Page.