Capturing the frugal beauty of complex natural tessallations

November 21, 2018

Surface tessellations are an arrangement of shapes which are tightly fitted, and form repeat patterns on a surface without overlapping. Imagine the pattern of a giraffe's fur, the shell of a tortoise and the honeycomb of bees -- all form natural tessellations. Mimicking these natural designs computationally is a complex, multi-disciplinary problem. A global team of computer scientists has developed a new, alternate model for replicating these intricate surface designs, veering away from classical, multi-step approaches to a more efficient, streamlined algorithm.

"When we look at how natural tessellation occurs in nature, the individual cells grow simultaneously, and each individual cell does not necessarily know who are its neighboring cells nor their location or coordinates," explains lead author of the work, Rhaleb Zayer, researcher at Max Planck Institute for Informatics in Saarbrücken, Germany. Cells represent the shape or tiles that comprise intricate tessellation patterns. "To capture this behavior, we need to adopt an intrinsic view of the problem and depart from the widely adopted extrinsic perspective which requires full knowledge of all individual cell interactions and locations."

Typically, researchers have turned to the Voronoi model to mimic such repeat surface patterns. In mathematics, A Voronoi diagram partitions planes in a pattern based on the distances between points. Efforts at extending the same idea to surfaces are hampered by the extensive costs of accurate distance measurements, bookkeeping and intersection computations.

In this new work, researchers simplify the creation of natural tessellations on surface meshes by dropping the assumption that regions need to be separated by lines. Instead, they have developed a method that takes into account region boundaries in the pattern as narrow bands, which are not necessarily straight, and model the partition as a set of smooth functions layered over the surface. Their method relies mainly on basic sparse linear algebra kernels, i.e. multiplication and addition, readily available, as they are the cornerstone of modern numerical computing.

"In this way, we provide small, concise, humanly readable and most importantly, platform-independent parallel code," notes Zayer.

"Observing the progress made in parallelizing existing serial Voronoi diagram codes over the last two decades, the performance gains achieved by our proposed method are very considerable," adds Markus Steinberger, coauthor of the work and an assistant professor at Graz University of Technology in Austria.

Zayer, Steinberger and their collaborators, which include Hans-Peter Seidel at Max Planck Institute for Informatics, and Daniel Mlakar at Graz University of Technology, will present their novel method at SIGGRAPH Asia 2018 in Tokyo 4 December to 7 December. The annual conference features the most respected technical and creative members in the field of computer graphics and interactive techniques, and showcases leading edge research in science, art, gaming and animation, among other sectors.

In their paper, "Layered Fields for Natural Tessellations on Surfaces," the authors successfully demonstrate their new method on several large-scale test cases beyond the capabilities of state-of-the-art. They were able to show that their method is applicable to highly detailed models, such as the 3D Model of the famed pilot Amelia Earhart's flight suit, encompassing ten million facets. Tessellations on the scan of the highly ornamented historic Pergolesi Side Chair showcase 30 million facets processed fully and efficiently on a single modern graphics-processing unit, aka, GPU. Despite the simplicity of the algorithm, the researchers say their solution proved to be comprehensive with minimal requirements.

In future work, Zayer and team hope to add the function of interactively editing tessellations using their framework. This feature could be aimed at designers and architects new to 3D printing applications and modeling. The researchers also intend to extend this work to higher dimensions and to the treatment of other metrics.
-end-
About SIGGRAPH Asia 2018

The 11th ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia (SIGGRAPH Asia 2018) will be held in Tokyo, Japan at the Tokyo International Forum from 4 to 7 December 2018. The annual event held in Asia attracts the most respected technical and creative people from all over the world who are excited by research, science, art, animation, gaming, interactivity, education and emerging technologies.

The four-day conference will include a diverse range of juried programs, such as the Art Gallery, Computer Animation Festival, Courses, Emerging Technologies, Posters, Technical Briefs, Technical Papers and Virtual & Augmented Reality. A three-day exhibition held from 5 to 7 December 2018 will offer a business platform for industry players to market their innovative products and services to the computer graphics and interactive techniques professionals and enthusiasts from Asia and beyond. For more information, please visit http://sa2018.siggraph.org. Find us on: Facebook, Twitter, Instagram and YouTube with the official event hashtag, #SIGGRAPHAsia.

About ACM SIGGRAPH

The ACM Special Interest Group on Computer Graphics and Interactive Techniques is an interdisciplinary community interested in research, technology, and applications in computer graphics and interactive techniques. Members include researchers, developers, and users from the technical, academic, business, and art communities. ACM SIGGRAPH enriches the computer graphics and interactive techniques community year-round through its conferences, global network of professional and student chapters, publications, and educational activities. For more information, please visit http://www.siggraph.org.

Association for Computing Machinery

Related Computer Scientists Articles from Brightsurf:

How computer scientists and marketers can create a better CX with AI
A failure to incorporate behavioral insight into technological developments may undermine consumers' experiences with AI.

UCLA computer scientists set benchmarks to optimize quantum computer performance
Two UCLA computer scientists have shown that existing compilers, which tell quantum computers how to use their circuits to execute quantum programs, inhibit the computers' ability to achieve optimal performance.

Computer vision helps SLAC scientists study lithium ion batteries
New machine learning methods bring insights into how lithium ion batteries degrade, and show it's more complicated than many thought.

Computer scientists create a 'laboratory' to improve streaming video
The quality of internet TV depends on algorithms that minimize glitches and stalls.

Computer scientists' new tool fools hackers into sharing keys for better cybersecurity
Instead of blocking hackers, a new cybersecurity defense approach developed by University of Texas at Dallas computer scientists actually welcomes them.

Computer scientists develop new tool that generates videos from themed text
A global team of computer scientists, from Tsinghua and Beihang Universities in China, Harvard University in the US and IDC Herzliya in Israel, have developed ''Write-A-Video'', a new tool that generates videos from themed text.

New research brings scientists one step closer to a fully functioning quantum computer
Quantum computing has the potential to revolutionize technology, medicine, and science by providing faster and more efficient processors, sensors, and communication devices.

Computer scientists predict lightning and thunder with the help of artificial intelligence
Together with Germany's National Meteorological Service, the Deutscher Wetterdienst, computer science professor Jens Dittrich and his doctoral student Christian Schön from Saarland University are working on a system that is supposed to predict local thunderstorms more precisely than before.

Computer scientists create programmable self-assembling DNA
Computer scientists at UC Davis, Maynooth University in Ireland and the California Institute of Technology have created DNA molecules that can self-assemble into patterns essentially by running their own program.

Computer scientists create reprogrammable molecular computing system
Caltech researchers have designed self-assembling DNA molecules with unprecedented reprogrammability.

Read More: Computer Scientists News and Computer Scientists Current Events
Brightsurf.com is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com.