“An efficient level set toolkit for visual effects”
Conference:
Type(s):
Title:
- An efficient level set toolkit for visual effects
Session/Category Title: Splashing in Pipelines
Presenter(s)/Author(s):
Abstract:
Over the past couple of years we have developed various tools for visual effects in movie production based on level sets. While most VFX studios, and even some commercial software packages, have had level sets tools for several years we are, to the best of our knowledge, among the first to employ highly optimized data structures. These compact and very fast volumetric data structures allow us to work at unprecedented grid resolutions, which in turn opens the door to new applications. The last two years we have demonstrated this with level set tools specifically developed for fluid surfacing in “Pirates of the Caribbean: At World’s End” [Museth et al. 2007] and fragmentation in “The Mummy: Tomb Of The Dragon Emperor” [Museth and Clive 2008], However, none of these past presentations actually discussed the details of the novel data structure, DB-Grid, which forms the very backbone of our level set implementations. The same was true for many of our fundamental level set tools, like the fast and robust scan-converter of self-intersecting polygonal meshes. These are exactly the topics of the current presentation. We will also show how these tools are tightly integrated with Houdini.
References:
1. Houston, B., Nielsen, M., Batty, C., Nilsson, O., and Museth, K. 2006. Hierarchical RLE Level Set: A Compact and Versatile Deformable Surface Representation. ACM Transactions on Graphics 25, 1, 3–24.
2. Museth, K., and Clive, M. 2008. Cracktastic: fast 3d fragmentation in “the mummy: Tomb of the dragon emperor”. In SIGGRAPH ’08: ACM SIGGRAPH 2008 talks, ACM, New York, NY, USA.
3. Museth, K., Clive, M., and Zafar, N. B. 2007. Blobtacular: Surfacing particle systems in “pirates of the caribbean 3”. In SIGGRAPH ’07: ACM SIGGRAPH 2007 talks, ACM, New York, NY, USA.
4. Nielsen, M. B., and Museth, K. 2006. Dynamic Tubular Grid: An efficient data structure and algorithms for high resolution level sets. Journal of Scientific Computing 26, 3, 261–299.
