This is one of the very few modeling techniques that can be used for various modeling scenarios. This modeling technique is used both in organic and inorganic modeling. The basic principle behind this modeling technique is as follows. The 3D artist draws the shape of the model he wants to create using the polygon tool. Then the polygon will get subdivided and extruded, so that the model gets the 3D form. This process of subdividing and extruding continues until the whole model is completed. In this modeling technique the artist in fact creates a polygonal mesh himself. But this polygonal model will have the minimal number of polygons only, and it will be far from a smooth and perfect figure. After this stage, the artist would apply a smoothing modifier on this polygon model, which converts the model into a realistic and smooth looking character.

This is a hybrid modeling technique, which inherits the merits of both NURBS modeling and Polygonal Modeling. It also offers a host of other enhancements. In sub-divisional modeling, the artist usually starts with a polygonal model itself. The polygonal model that has been already created is converted into a sub-divisional model. The sub-divisional model has control points and lattices that can be pulled and moved, to alter the original polygonal model that lies beneath. This modeling technique also permits the artist to select the level of refinement to be done in a particular area. Hence those areas requiring higher detail can be assigned with more control points and higher no: of vertices, whereas the other areas will remain with a base vertex density. This gives significant performance boost.