Driving Object Deformations from Internal Physical Processes Zeki Melek ∗ Department of Computer Science Texas A&M University John Keyser † Department of Computer Science Texas A&M University Figure 1: A bending match while burning. Abstract In this paper we present a method for deforming objects for graph- ics applications, based on the results of internal physical simula- tions. As driving examples, we describe in detail methods for sim- ulating the bending of burning matches, and the crumpling of burn- ing paper. In these cases, the small-scale changes in a chemical process result in large-scale deformations of the given object. We propose the use of a free form deformation to model such large- scale deformations. Changing object properties are mapped onto the edges of a proxy object, which is then modified by treating the edges as springs. This proxy object then serves as a control struc- ture for defining the deformation of the underlying object. The re- sults we present are fast, controllable, and visually plausible. CR Categories: I.3.5 [Computer Graphics]: Computational Ge- ometry and Object Modeling—Physically Based Modeling Keywords: physically based modeling, deformation 1 Introduction In the push for greater realism in computer graphics applications, complex physical simulations are playing a larger and larger role. From real-life experience, users are often familiar with the physical phenomena being simulated, and different simulations have vary- ing success at replicating the details of a physical process. Such details can make the difference between a believable virtual world that draws the user in and a jarring environment that destroys any sense of presence. Many complex physical simulations require modeling of a number of different phenomena. Depending on the particular application, these phenomena can require different levels of accuracy in order to create an overall visually plausible result. Though an “ideal” ∗ e-mail: melekzek@tamu.edu † e-mail:keyser@cs.tamu.edu simulation might accurately simulate all details of all physical phe- nomena, this is usually impractical, given constraints on time, pro- cessing capability, and even underlying knowledge of the physical process. Instead, what is usually done in the computer graphics community is to simplify the physical model, use a simpler simu- lation, and eliminate certain secondary effects in order to achieve a plausible result in reasonable time. This is the case for all simula- tions, but it is particularly magnified in interactive applications. The goal of our work is to develop a method for efficiently modeling certain secondary effects in physical simulations, thereby increas- ing visual plausibility of the overall simulation for only a reasonable cost in efficiency. We do this by attempting to model the large-scale effects of certain physical processes, rather than spending a dispro- portionate amount of computation on a minor yet potentially com- plex phenomenon. In particular, in this paper we propose a way of approximating larger-scale deformations of objects guided by the results of a simulated physical process. An example of such complex physical processes is burning objects. To simulate a burning object, the combustion reaction, heat distribu- tion, fuel consumption, and even object shape must be modeled and changed over time. The pyrolysis process, where an object releases combustible gases, causes decomposition and additional structural changes in burning objects. Although these structural effects are usually minor, some create a dominant deformation on burning ob- jects. One such structural change is caused by microscopic contrac- tion of fibers within a burning object. This results in effects that are quite noticeable at the macroscopic level, such as the way matches bend when burnt, and the way burning paper tends to crumple. We can simulate the combustion reaction, the object catching fire and burning, and even the decomposition of the object as it burns, but unless we model the fiber contraction, the simulated matches will not behave like actual burning matches (fig. 2). In this paper we present a free form deformation (FFD) based method for approximating large-scale deformations due to smaller- scale physical simulations. We also show specific examples of how this approach can be applied to represent the deformation of burn- ing objects according to the changing object properties encountered during the simulation. The major contributions of this paper are • We present a framework for creating deformations guided by physical simulations. This includes: – Defining a proxy object. The deformations are simu-