vriphys09
Permanent URI for this collection
Browse
Browsing vriphys09 by Subject "Categories and Subject Descriptors (according to ACM CCS): Computer Graphics [I.3.5]: Computational Geometry and Object Modeling - Physically based modeling, Computer Graphics [I.3.7]: Three-Dimensional Graphics and Realism - Animation"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Optimized Impulse-Based Dynamic Simulation(The Eurographics Association, 2009) Bayer, Daniel; Diziol, Raphael; Bender, Jan; Hartmut Prautzsch and Alfred Schmitt and Jan Bender and Matthias TeschnerThe impulse-based dynamic simulation is a recent method to compute physically based simulations. It supports the simulation of rigid-bodies and particles connected by all kinds of implicit constraints. In recent years the impulse-based dynamic simulation has been more and more used to simulate deformable bodies as well. These simulations create new requirements for the runtime of the method because very large systems of connected particles have to be simulated to get results of high quality. In this paper several runtime optimizations for the impulse-based dynamic simulation are presented. They allow to compute the same simulations at a fraction of time needed for the original method. Therefore, larger systems or simulations with increased accuracy can be simulated in realtime.Item Simulating Almost Incompressible Deformable Objects(The Eurographics Association, 2009) Diziol, Raphael; Bayer, Daniel; Bender, Jan; Hartmut Prautzsch and Alfred Schmitt and Jan Bender and Matthias TeschnerWe present a new method for simulating almost incompressible deformable objects. A tetrahedral model is used to represent and restore the volume during the simulation. A new constraint, which computes impulses in the one-ring of each vertex of the tetrahedral model, is used in order to conserve the initial volume. With different parameters, the presented method can handle a large variety of different deformation behaviors, ranging from stiff to large deformations and even plastic deformations. The algorithm is easy to implement and reduces the volume error to less than 1 percent in most situations, even when large deformations are applied.