A scientific Oscar goes to Stanford
On February 9, 2008, the Academy of Motion Picture Arts and Sciences will give its ten Scientific and Technical Academy Awards for the year. One of them will go to a professor of computer science at Stanford University who worked with scientists from Industrial Light & Magic. They'll receive this 'Oscar' for their work about cyber-fluids. They've started to work several years ago for the development of the female liquid terminator in Terminator 3. This technology also was used to model the sinking ship in Poseidon and in the two latest Pirates of the Caribbean movies. But read more...
You can see above several computer-generated scenes showing off the fluid simulation technology which will receive a scientific Academy Award this week at The Beverly Wilshire. (Credit: Ron Fedkiw and his team, Stanford University) But don't miss this short movie (16 seconds) showing the waves trying to invade the lighthouse shown above.
This research has been led by Ron Fedkiw, Associate Professor of Computer Science, and Stanford researchers Jerry Talton and Nipun Kwatra. But Fedkiw, who has also worked for Industrial Light & Magic (ILM) before, collaborated on this project with ILM scientists Nick Rasmussen and Frank Losasso Petterson, who will share the award with him.
But why these scientists have been honored by the Academy of Motion Picture Arts and Sciences? Let's read the Stanford Report. "Computer graphics experts typically have used particles and complex blobs to represent water, but these can give rise to unrealistically lumpy or grainy surfaces. Alternatively, they have used a technique called 'the level set method' that gives a smooth surface representation, but some water is 'under-resolved' and simply disappears when it breaks down into small volumes, as in a crashing wave."
So what are the new techniques developed by Fedkiw and his fellow researchers? "The key innovation behind Fedkiw and former doctoral student Douglas Enright's novel 'particle level set method' was to mix the use of particles and level sets so that studios could maintain smooth surfaces wherever possible and still keep all the fluid via the particle representation. 'As an added bonus, the method automatically generates spray particles and bubbles in under-resolved regions where the level set [method] loses mass and volume,' Fedkiw said. "
For more information, this research work will be published in a future issue of IEEE Transactions on Visualization & Computer Graphics under the name "Two-way Coupled SPH and Particle Level Set Fluid Simulation." Here is the beginning of the abstract. "Grid-based methods have difficulty resolving features on or below the scale of the underlying grid. Although adaptive methods (e.g. RLE, octrees) can alleviate this to some degree, separate techniques are still required for simulating small-scale phenomena such as spray and foam, especially since these more diffuse materials typically behave quite differently than their denser counterparts."
And here is a description of the approach of the researchers. "In this paper, we propose a two-way coupled simulation framework that uses the particle level set method to efficiently model dense liquid volumes and a smoothed particle hydrodynamics (SPH) method to simulate diffuse regions such as sprays. Our novel SPH method allows us to simulate both dense and diffuse water volumes, fully incorporates the particles that are automatically generated by the particle level set method in under-resolved regions, and allows for two way mixing between dense SPH volumes and grid-based liquid representations."
If you want to learn more about this method to create computer-generated graphics, here is a link to the full paper as it has been submitted to IEEE TVCG (PDF format, 7 pages, 12.0 MB). The above illustration has been extracted from this document.
Finally, other projects received an Academy Award. Here is a link to the 10 scientific and technical winners of the Scientific and Engineering Awards given by the Academy of Motion Picture Arts and Sciences.
Sources: David Orenstein, Stanford Report, January 16, 2008; and various websites
You'll find related stories by following the links below.