Graphics Reference
In-Depth Information
FIGURE 9.32
Hair modeled as strips of strands [ 37 ] .
ABC
FIGURE 9.33
Hair modeled using multiple levels of detail [ 65 ].
To provide both the computational savings of strand clusters as well as the realism of animating
individual strands, a multiple level of detail approach can be used ( Figure 9.33 , Color Plate 6) [ 65 ]
and, to provide more flexibility, adaptive grouping can be used [ 64 ] . Other approaches include model-
ing hair as a thin shell [ 34 ] and as a continuum [ 24 ].
9.5 Chapter summary
The human figure is an interesting and complex form. It has a uniform structure but contains infinite
variety. As an articulated rigid structure, it contains many DOFs, but its surface is deformable. Model-
ing and animating hair in any detail is also enormously complex. Moreover, the constant collision and
sliding of cloth on the surface of the body represents significant computational complexity.
One of the things that makes human motion so challenging is that no single motion can be identified
as correct human motion. Human motion varies from individual to individual for a number of reasons,
but it is still recognizable as reasonable human motion, and slight variations can seem odd or awkward.
Research in computer animation is just starting to delve into the nuances of human motion. Many of
these nuances vary from person to person but, at the same time, are very consistent for a particular
person, for a particular ethnic group, for a particular age group, for a particular weight group, for a
particular emotional state, and so on. Computer animation is only beginning to analyze, record, and
 
Search WWH ::




Custom Search