Key points are not available for this paper at this time.
The retinal image of a visual scene consists of a two-dimensional, continuous distribution of gray levels. Before the visual system can identify particular figures or objects, it needs to determine which of the various luminance values belong to individual objects or to the embedding background. Some grouping has to be performed in order to associate these luminance distributions with contours, to associate particular contours with a single object, and to segregate objects with overlapping contours from each other and from the background.
Singer et al. (Mon,) studied this question.