As stated just previously, multiple polygons are how we define objects and shapes. This collection of polygons is referred to as a mesh. A mesh is effectively a wireframe modelling of the object, that also shows all the individual polygons that make it up. Polygons can then be grouped into categories such as strips, fans, and so on. This allows for a more efficient data storage, in that shared vertices between polygons don’t have to be redefined. Due to the discrete nature of computer data versus the continuous nature of real-world objects and measurements, meshes and polygons are always a rough approximation to real curves and surfaces.

Meshes are very useful for visualising data and generating accurate models, but when it comes to data processing, are still just a collection of vertices that need to be processed by the system. Any changes applied to the object is actually that change applied to all the individual vertices within that object’s mesh. These changes are referred to as transformations.