More than skin deep : physical modeling of facial tissue

This thesis describes a computer system for animating human facial tissue using a mathematical model of the stresses and strains caused by deformation of the tissue. The model is based on networks, the nodes of which represent reference points within the tissue, and the arcs of which are spring-like constraints which model the mechanical behavior of the tissue between the reference points. The networks can be assembled to approximate the anatomy of facial tissue. The system simulates the action of layers of soft tissue as they interact with underlying hard tissue, internal muscle forces, and external forces such as gravity. The ability to combine the effect of these forces has been missing in previous animation models of facial tissue. The system is part of a project to develop real-time animation systems in which the simulated objects respond as if they were real objects. An application of the physical model of facial tissue and real-time manipulation of simulated objects includes a simulator for planning surgery and training plastic surgeons. The techniques described here can also be used to simulate facial tissue for computer animated human figures and create more realistic facial expressions than available in previous animation models. The facial tissue simulation system comprises several modules of an interactive simulation environment, called bolio, which is also briefly described here. Thesis Supervisor: David L. Zeltzer Title: Associate Professor of Computer Graphics