Carnegie Mellon University: Computer graphic artists who produce computer-animated movies and games spend much time creating subtle movements such as expressions on faces, gesticulations on bodies and the draping of clothes. A new way of modeling these dynamic objects, developed by researchers at Carnegie Mellon University, Disney Research, Pittsburgh, and the LUMS School of Science and Engineering in Pakistan, could greatly simplify this editing process.
6 comments:
You can have whatever opinion you so choose of Disney (I know many sentiments are negative.) And, for that matter, any opinion you want of Carnegie Mellon. But, in spite of all that, it is undeniable that what happens at this school is groundbreaking. The amount of software and technologies produced here rivals that of any other institution in the nation. And, even beyond the walls of Purnell (arguable even more so,) creativity is rampant. Playing off of that, Disney is unrivaled in the business of making magic. Seeing the two as a partnership is not only cool, but seems a good fit as well.
The Disney Research center at Carnegie Mellon continues to pump out some really interesting projects. I had the chance to be a part of one study involving the natural appearance of eye blinks, which went forward to being published and then the techniques being used in Disney and Pixar animations from that point on. What I found really fascinating about this project is that rather than assuming that breaking down the problem (motion tracking) into its relative parts (space and time), thinking of the movement holistically (and in some ways, much more naturally) helped to reduce the overall effort needed. There's a lesson in there of keeping an eye on the bigger picture and that sometimes a solution can be found more easily by bringing together two divergent processes, because you can then drop the common ground between them and move forward more effectively and efficiently. "Collaborative art", indeed.
Although I feel like I didn't totally grasp a lot of the explanation of what was actually discovered here, from what I can tell this seems like really exciting work for computer animation. The technology that exists for motion right now, at least on most platforms that we're familiar with like After Effects, works only up until a point and definitely, like the article says, still takes hours to complete a simple sequence. If researchers here have found a more realistic way to program motion, especially human motion, they are in for a lot of due celebrity. The algorithms that animators have to come up with in order to mimic human motion are insane and enormously complex. I know that just to get the hair in Brave right they needed to develop two completely separate algorithms which took months and months. I would love to be able to see a demonstration of what they've come up with on the other side of campus.
I can't say that I completely understand how this new modeling system works, but given the quote in the last paragraph of this article from a senior scientist at Disney Research Pittsburgh, this must be a pretty big deal. What I love is the fact that this modeling method is actually simplifying the process instead of over-complicating. It seems like this stream-lined process can save people in many fields quite a bit of time, money, and stress when it comes to editing animated sequences. I am always humbled when I hear about the amazing accomplishments of our fellow CMU students, and I am especially thrilled to know that CMU and Disney are doing such awesome things together. How can you beat that combination? This is a scientific development that may not necessarily save lives, but that will hopefully make some peoples' lives easier, and that's what we should be trying to do, right? What I really want to say is that I could never do this but I'm so happy that the people who can do it are here at CMU, making a difference one modeling method at a time.
One of the most interesting parts of this to me is the fact that one of the major players in it is at a University in Pakistan. Carnegie Mellon and Disney are at the top of the curve in collaborating with the most brilliant people from across the globe. CMU has campuses in multiple countries, and this is another facet of that important interaction. I would love to see how we, as Drama artists, can further collaborate to develop new technology, new art forms, and new works with people from around the world. I expect we will continue to see fascinating releases from the Disney Research Pittsburgh branch, and I'm excited to see what they are!
I think this is pretty amazing. First how any different people are working together with such great success. Personally I enjoy what they have been able to do with computer graphics in movies and tv. Seems like they truly have made it much simpler and also more efficient to produce more accurate facial modeling. I do wonder if this will increase speed of production. This sounds amazing I would love to know how this ends up effecting the industry at large.
Post a Comment