Cumhur Ozan Çetinaslan recently published an important conference paper on “ESPEFs: Exponential Spring Potential Energy Functions for Simulating Deformable Objects” in the 14th ACM SIGGRAPH Conference on Motion, Interaction, and Games.

Cumhur is a postdoctoral researcher in IT-Porto. He conducts research in the topics of Visual Simulation, Computational Animation, and Computer Graphics.

Paper synopsis:

Physically-based visual simulation of natural phenomena is a longstanding research subject of computer graphics. Although there exist many methods on the subject, Position-based Dynamics (PBD) [Müller et al. 2006] is the most popular method that is used for physics-based simulations in the most common 3D modeling and animation software. The popularity of PBD is explained by its stability, efficiency, and simplicity. Therefore, many different natural phenomena, such as cloth, fur, hair, fluid, rigid bodies, deformable objects, skinning, were simulated by employing PBD technique over the years. In 2016, Extended Position-based Dynamics (XPBD) was proposed by Macklin et al. They improved the numerical solver of the original PBD with additional parameters. Therefore, adjusting many parameters for the desired simulations becomes a cumbersome and time-consuming task with many trial-and-error tests. In our paper, ESPEFs: Exponential Spring Potential Energy Functions for Simulating Deformable Objects, we reduce the process of parameter adjustments and provide more vivid motion during the physics-based simulation of elastic materials. Our method reformulates the well-known spring potential energy functions on an exponential basis and enriches the hyperelastic behavior of the deformable models. The resulting approach is simple, stable, interactive, and produces visually pleasing results compared to the existing methods.

Further information on the conference webpage: https://mig2021.inria.fr/

See also the illustrative video presentation (above).