Visual objects and data objects

Visual objects are central to the human understanding of the higher levels of visual processing. In a sense, the object can be thought of as the point at which the image becomes thought. Objects are units of cognition and things that are recognized in the environment. There is strong evidence to support both viewpoint-dependent recognition of objects and the theory that the brain creates 3D structural models of objects. Visual processing of objects is very different from the massive processing of low-level features. Only a very small number of complex visual objects, perhaps only one or two, can be held in mind at any given time. This makes it difficult to find novel patterns that are distributed over multiple objects. However, there is a kind of parallelism in object perception. Although only one visual object may be processed at a time, all the features of that object are processed together. This makes the object display the most powerful way of grouping disparate data elements together. Such a strong grouping effect may not always be desirable; it may inhibit the perception of patterns that are distributed across multiple objects. This chapter explains number of ways in which different spatial variables interact to help us recognize objects, their structures, and their surfaces.