Showing posts with label computer graphics. Show all posts
Showing posts with label computer graphics. Show all posts

Apr 1, 2013

The Uncanny Valley is Here! Activision's real-time character demo is chillingly real. (Not an April 1st joke.)

Up with Activision in the Uncanny Valley

I first heard the term 'uncanny valley' about eight or nine years ago when I was taking a 3D-modeling class.  At that time, the technology available was not close to reaching this valley - where robots or computer-generated characters are so real that they are almost repulsive.

A lot has changed over the years.

The following video, recently featured by Activision Blizzard during the 2013 Game Developer Conference, has attracted much attention in just a few days, partly because it is so real.



Although I noticed that a little more work needs to be done with the teeth, I was impressed. I liked the quality of the eye shaders that were used in the creation of this demo.  Examples of faces created with this feature turned on and off can be found on Jorge Jimenez' blog.  Jorge's slides from a 2012 course offered during SIGGRAPH provide additional information.

Computer processers have become powerful enough to handle quite a lot of processing, and the tech world has been spreading the word. Below is a presentation by Jen-Hsun Huang, CEO of Nvidia, about the company's work simulate the human face, touching on the 'uncanny valley':


Although the use of this technology to create characters, realistic on many levels, seems to be a bit creepy, it might be OK after some refinement.  There are a few questions that remain unanswered.   What would be the impact on children or teens who might spend many hours each week playing games with such realistic characters?  I'd hate to have a nightmare featuring one of these guys!

I think that this technology might have some potential for use in serious games and simulations, such as preparing emergency workers to handle a variety of realistic scenarios. Games with realistic digital characters, capable of generating a range of facial expressions, might be useful to support the learning of social interaction skills among young people with autism spectrum disorders.


RELATED/SOMEWHAT RELATED
Is It Real?  With New Technology Has Activision Crossed the 'Uncanny Valley'?
Eyder Peralta, the two-way, NPR, 3/28/13
Activision Reveals Animated Human That Looks So Real, It's Uncanny
Charlie White, Mashable, 3/28/13
Karl F. MacDorman's Writings (some focus on the uncanny valley)
Advances in Real-Time Rendering in Games Course (SIGGRAPH2012)
Separable Subsurface Scattering and Photorealistic Eyes Rendering (pptx)
Jorge Jimenez, Presenter, SIGGRAPH 2012)
Next Generation of Character Rendering Teaser (pptx)
Jimenez and Team
Crossing the 'uncanny valley': Nvidia's Faceworks renders realistic human faces
Dean Takahashi, VB/Gamesbeat, 3/18/13
Real-Time Realistic Skin Translucency
Jimenez, J., Whealan, D., Sundstedt, V., Gutierrez, D IEEE Computer Graphics and Applications, 2010
Exploring the Uncanny Valley Research Website (Indiana University School of Informatics)
Gaze-based Interaction for Virtual Environments pdf
Jimenez, J., Gutierrez, D., Latorre, P.  Journal of Universal Computer Science
Mori, Masahiro (1970). Bukimi no tani [the uncanny valley] (K. F. MacDorman & T. Minato, Trans.). Energy, 7(4), 33-35. 
2013 GPU Technology Conference Keynote Presentations

Jun 1, 2011

Fun, weird, and a little scary: "Baaa" and "Cows&Cows&Cows" - video clips by Cyriak Harris for a quick smile!

Cyriak Harris is a freelance animator, illustrator, and graphic designer who lives near Brighton, UK. He uses Adobe software to create his unique animations.


Baaa "Experiments in ovine geometry",  featuring self-replicating sheep:



"surreal bovine choreography. No cows were harmed during the making of this video, though their future prospects probably aren't as optimistic. Software used: adobe after effects"



I liked the fractal hand experiment:

Here is an interesting music video experiment:




RELATED
Cyriak's Blog
Cyriak's Website
Cyriak's Wikipedia Page

Aug 20, 2008

The Hidden Geometry of Hair: Computer Generated "Hollywood Hair" Unveiled at SIGGRAPH 2008








I came across a link to an article in the UCSD School of Engineering News today and just had to post it.

"HOLLYWOOD HAIR IS CAPTURED AT LAST: DETAILS IN SIGGRAPH 2008 PAPER"

Technology now exists that will enable animators, video-game makers, and film-makers the opportunity to capture images of real hairstyles that look and behave realistically. Below is a series of side-by-side comparisons of computer generated heads of hair and the real thing.

From the article:

"By determining the orientations of individual hairs, the researchers can realistically estimate how the hairstyle will shine no matter what angle the light is coming from. “You can’t just blend the highlights from two different angles to get a realistic highlight for a point in between,” said Chang.
“Instead of blending existing highlights, we create new ones.”


....One possible extension of this work: making an animated character’s hair realistically blow in the wind. This is possible because the researchers also developed a way to calculate what individual hair fibers are doing between the hairstyle surface and the scalp. They call this finding the “hidden geometry” of hair.

“Our method produces strands attached to the scalp that enable animation. In contrast, existing approaches retrieve only the visible hair layer,” the authors write in their SIGGRAPH 2008 paper. An animation of a hairstyle is available as a download from the “hair photobooth” Web site created by Sylvain Paris: http://people.csail.mit.edu/sparis/publi/2008/siggraphHair/

What we need is a real-life hair photobooth. Have a bad hair day? No problem. Walk in, "zap!", and walk out with real Hollywood hair. Maybe the hair scientists are working on it....


SIGGRAPH 2008 Paper citation: “Hair Photobooth: Geometric and Photometric Acquisition of Real Hairstyles,” by Sylvain Paris and Wojciech Matusik from Adobe Systems, Inc., Will Chang, Wojciech Jarosz and Matthias Zwicker from University of California, San Diego; and Oleg I. Kozhushnyan and Frédo Durand from Massachusetts Institute of Technology