Model

In the language of 3-D graphics, a model is a data file that contains the information needed to view or "render" a 3-D object. This information includes two types of information:

1. The geometry--the shape--of the object.

2. The surface attributes of the object--meaning data that allows the object to be properly colored so that it looks like it is made of some kind of material (e.g. metal, glass, wood, plastic, etc.)

This glossary entry will consider only the first type--the geometry information in the model.

The geometry information in a model defines the surfaces of the object as a list of flat polygons that share common sides and vertices (corner points). The model therefore describes a mesh. (There are other, more sophisticated ways of describing surface geometry--but these are not important for the beginner.) Note that if the polygons don't share common sides and vertices, there will be breaks in the model, and it will not describe the continuous surface of a 3-D object.

The typical model divides its geometry information into two parts. First, there is a list of points representing the vertices of the polygons as (x,y,z) coordinates.

Vertex List:

Point 1 (x,y,z)

Point 2 (x,y,z)

Point 3 (x,y,z)

Point 4 (x,y,z)

......

Point 10 (x,y,z)

Second, this list of points is used to describe polygons.

Polygon List:

Polygon 1 (Point 1, Point 2, Point 3, Point 4)

Polygon 2 (Point 4, Point 5, Point 7, Point 9)

...

Polygon 6 (Point 3, Point 5, Point 6, Point10)

This approach has the advantage of keeping polygons connected. If the coordinates of any point are changed in the Vertex List, the change is necessarily reflected in every polygon on the Polygon List that uses that point as a vertex.

There is no better way of understanding the basics of models than to play with VRML files. VRML files use the very file structure described here. By bouncing back and forth between the text of the VRML file in a word processor, and the rendering of that file in a VRML browser, you can quickly grasp the connection between the contents of the file and the 3-D image.

Comments are welcome