Texture Segregation, Surface Representation, and Figure-ground Separation
MetadataShow full item record
A widespread view is that most of texture segregation can be accounted for by differences in the spatial frequency content of texture regions. Evidence from both psychophysical and physiological studies indicate, however, that beyond these early filtering stages,there are stages of 3-D boundary segmentation and surface representation that are used to segregate textures. Chromatic segregation of element-arrangement patterns as studied by Beck and colleagues - cannot be completely explained by the filtering mechanisms previously employed to account for achromatic segregation. An element arrangement pattern is composed of two types of elements that are arranged differently in different image regions (e.g., vertically on top and diagonally on bottom). FACADE theory mechanisms that have previously been used to explain data about 3-D vision and figure-ground separation are here used to simulate chromatic texture segregation data, in eluding data with equiluminant elements on dark or light homogenous backgrounds, or backgrounds composed of vertical and horizontal dark or light stripes, or horizontal notched stripes. These data include the fact that segregation of patterns composed of red and blue squares decreases with inereasing luminance of the interspaces. Asymmetric segregation properties under 3-D viewing conditions with the cquiluminant element;; dose or far arc abo simulated. Two key model properties arc a spatial impenetrability property that inhibits boundary grouping across regions with noncolinear texture elements, and a boundary-surface consistency property that uses feedback between boundary and surface representations to eliminate spurious boundary groupings and separate figures from their backgrounds.