Figure

The hierarchical organisation of the human figure (from Marr & Nishihara, 1978) at various levels: (a) axis of the whole body; (b) axes at the level of arms, legs, and head; (c) arm divided into upper and lower arm; (d) a lower arm with separate hand; and (e) the palm and fingers of a hand.

• Scope and uniqueness: "scope" is the extent to which the representation is applicable to all the shapes in a given category, and "uniqueness" means that all the different views of an object produce the same standard representation.

• Stability and sensitivity: "stability" indicates that a representation incorporates the similarities among objects, and "sensitivity" means it incorporates salient differences.

Marr and Nishihara proposed that the primitive units for describing objects should be cylinders having a major axis. These primitive units are hierarchically organised, with high-level units providing information about object shape and lowlevel units providing more detailed information. Why did Marr and Nishihara adopt this axis-based approach? They argued that the main axes of an object are usually easy to establish regardless of the viewing position, whereas other object characteristics (e.g., precise shape) are not.

We can illustrate Marr and Nishihara's theoretical approach by considering the hierarchical organisation of the human form (see Figure 4.4). The human form can be decomposed into a series of cylinders at different levels of generality. It was assumed that this overall 3-D description is stored in memory, and enables us to recognise appropriate visual stimuli as humans regardless of the angle of viewing. According to Marr and Nishihara (1978), object recognition involves matching the 3-D model representation constructed from a visual stimulus against a catalogue of 3-D model representations stored in memory. To do this, it is necessary to identify the major axes of the visual stimulus. Marr and Nishihara proposed that concavities (areas where the contour points into the object) are identified first. With the human form, for example, there is a concave area in each armpit. These concavities are used to divide the visual image into segments (e.g., arms; legs; torso; head). Finally, the main axis of each segment is found.

There are some advantages associated with this emphasis on concavities and axis-based representations. First, the identification of concavities plays an important role in object recognition. Consider, for example, the faces-goblet ambiguous figure (look back at Figure 2.2), which was studied by Hoffman and Richards

An outline of Biederman's recognition-by-components theory. Adapted from Biederman (1987).

Essentials of Human Physiology

Essentials of Human Physiology

This ebook provides an introductory explanation of the workings of the human body, with an effort to draw connections between the body systems and explain their interdependencies. A framework for the book is homeostasis and how the body maintains balance within each system. This is intended as a first introduction to physiology for a college-level course.

Get My Free Ebook


Post a comment