CMU School of Drama


Wednesday, September 28, 2011

Carnegie Mellon, Disney work to improve eye blink animation

The Tartan Online: A group of Carnegie Mellon researchers, in association with Disney Research of Pittsburgh, are bringing animations closer to reality by modeling accurate eye blinks.
Conventional systems that model eye blinks have always assumed them to be symmetric. In other words, during an eye blink, a person’s eyelids move down at the same rate that they move back up. While this may be a rational assumption to make, researchers’ high speed cameras a slightly different story: real human eyelids go down quickly during an eye blink, followed by a more gradual opening back up.

5 comments:

Pia Marchetti said...

I love animation, especially Disney. (My favorite Disney movie is Sleeping Beauty.) I even considered pursuing a degree in animation or film instead of theatre.
It's particularly interesting what constitues a realistic animation. When I've tried to make short animations on my own, I've struggled with what was necessary. I realized quickly that not every different position can be drawn separately. In fact, very few frames can combine to create what looks like a realistic motion. I think that's what this study showed. I wish I could've been a part of it.

Brian Rangell said...

I was a participant in this study when it was being performed at the Disney Research Labs. I was definitely able to see a difference between the symmetrical eyeblinks and the asymmetrical, more natural ones. For me, the time ratio between the eye closing and the eye opening was the most significant indicator of a "realistic" animated eyeblink. The blinks looked robotic and false when it was too symmetrical, while for some of the more realistic ones, I couldn't tell for the life of me if it was animation or a video recording. The mind is stunningly attuned to these minute details.

Robert said...

 
This is interesting that CMU is teaming up with Disney to work on eye blinking.  I would think that CMU would also be doing something way bigger then that with all the talent that CMU has.  I wonder how Disney will be using this information that CMU is gathering. I wonder if they will just use it to reprogram some of their current attractions or develop new ones.  It seems that some of the things that they are discovering are things that they may just have to put a delay or hold on some months to make it seem more realistic.  I also wonder how CMU is gathering the data, are they putting high speed cameras in front of people and telling them to blink?  IF they are doing that it seems that they are not getting a true blink.
 

Katherine Eboch said...

What amazing technology CMU and Disney are trying to develop. I love that CMU has such a strong connection to Disney when it comes to imagineering. I disagree with Robert about doing bigger things, this is a pretty cool project to be working on. It may not be the cure to cancer, but it is an advancement in the animation world. Hopefully something cool will come from this research like a new movie or ride animation.

David Beller said...

My question is... At what point do me stop trying to make animations look like real life and instead accept that they are completely different form of expression?

Going a step farther, where is the line between a movie considered to be an animation and one to use CGI? Are there really any movies that are completely animation free?

Neither of these thoughts/questions are accusatory, but just make you think about the fact that except for a very few exceptions, everything that we see has been processed by computers in some way.