May 3-7, 2005
Thesis Presentations
721 Broadway
at Waverly Place
4th Floor
South Elevators
New York, NY 10003
For additional
information email
itp.inquiries@nyu.edu
or call 212.998.1880
| Animate Pillow Project | ||
|---|---|---|
| Author(s): | phaedra riley | |
| Instructor: | Zurkow, Marina | |
| Class: | Final Project Seminar | |
| Keywords: | pillow, animation, animate, animatronic, robot, robotics, roboticist, social robots, kinetic sculpture, art, social, interactive | |
| By selectively focusing on the details of movement and gesture over true-to-life reproduction, animators create a stronger emotional bond with their audiences by allowing people the visual space to enrich the animation with their own experiences. | ||
| Photorealistic animation does not leave enough visual white space for people to project their imaginations into in order to fill in the discrepancies between what they see and what they experience as real. Because of this the audience is denied access to the emotional projection and anthropomorphism that effectively creates empathy in animation. By selectively focusing on details of movement and gesture, animators create a stronger emotional bond with their audience by allowing people to enrich the animation with their own experiences. Relying on this, applicants at Disney are asked to submit animations of a sack of flour expressing various emotions as part of their reel as a test of their abilties to take something so blank and featureless as a pillow and imbue it with life. Playing off this idea, I propose that effective empathy can be attained with animatronics using the same basic elements: anthropomorphism, emotional projection, and the gestures and motion that can be achieved with a pillow -- as opposed to the standard robotic approaches which feature either extreme realism in the appearance of an living creature, or focus on multiple degrees of facial articulation. | ||
| Personal Statement: | *i like to animate
*generally the only way i really feel like i can express myself in the world is through animation. film is too collaborative. actors are too complicated. *i like the repetitiveness of it - like clothwork or sewing or quiliting (i suppose) *its the only thing i can lose myself in the way i hear programmers describe programming fits *i came to itp & tried to stay away from animating. but then i\'d do a bit for a project, and it was always so satisfying that i\'m now convinced its what i should be doing in some form *i also came to itp b/c while its easy to digitize things, its still hard to make them tangible again. this project is an attempt at a tangible animation. | |
| Context: | *movie/special FX industry (stan winston, WETA)
*uncanny valley *robotics - commercial *robotics - research *art - painting(concerned with movement) *art - kinetic sculpture *etymology (robot, animation, animatronic) *The most accessible place for the term \"animatronics\" is in the movie and special effects industry. *Stan Winston Studio is a front runner in the field of special effects animatronics. They have a stated goal to make their work art, and while I admire their craftmanship, I feel that their work is still merely special effects puppets rather than stand-alone creations. *WETA is probably the most talked-about special effects house right now for their work on the Lord of the Rings trilogy and more specifically for me, the character, Gollum. They used cutting edge motion capture techniques to create an entirely digitally represented character. While I believe the Gollum character was technically interesting, I was never as moved by it as I expected to be, and after seeing an almost impromptu performance by the actor, Andy Serkis, which blew me away I realized that while Gollum\'s face was hand-animated, it was still less expressive and animated than the original actor\'s. I find that motion-capture and photo-realism is best when used as a special effects tool, rather than to create and animate a character. *I think this is because human vision is so finely tuned to motion that anything that seems normal but slightly off is psychologically unsettling to us. The Japanese robotician, Professor Masahiro Mori, did some research in this direction and developed what is referred to as The Uncanny Valley theory. Professor Mori tested subjects\' reactions towards synthetic creatures with varying levels of similarity to normal humans and found that generally empathy increased the more human-like the creatures were - until a critical point when a sudden drop from empathy to outright revulsion occurred. This happened mostly when the synthetics closely resembled humans, but were still distinguishable as inhuman; but as they became more normally human again high empathy levels ensued. He ran the tests based on both appearance and movement with similar results. If the movement and appearance are both right, then empathy occurs. But since it takes so much more time and effort to recreate something, rather than just use the real thing, I believe that photo-realism is valuable as a skill or technique, but should not be the goal of animation. *Because of the pitfall of the Uncanny Valley, much human interactive robotics are stylized such as Aibo, *Kismet, or the in-development Leonardo who was also designed by SWS for the Robotic Life Group\'s Sociable Robots Project at MIT. *Beyond robotics, I am looking at art and kinetic sculpture for inspiration. *As Giannalberto Bendazzi notes in his book, Cartoons: one hundred years of cinema animation, \"The search for movement, which had constantly marked the history of art, had become pressing after Impressionism, when paintings strove ever more to capture life itself, leaving static representations to photography.\" Cartoons pg 12 *Arthur Ganson\'s kinetic sculptures are so open and accessible, yet use complex mechanics to create simple movement, such as in this piece, \"Thinking Chair.\" *alex the parrot | |
| Audience: | *traditional art viewers - this is not a piece about technology nor any sort of technical achievement. its about how we perceive animation & our relationship to objects.
*animators - to explore nontraditional methods of animation. to consider animatronics as more than bad, cheap robotics or special effects. to re-connect with the simpler fundamentals of why animation works at all *roboticists - to remember that it is the emotional bond that keeps things out of the uncanny valley. | |
| User Scenario: | *pillow is isolated on a stand (white? lit from above?)
*pillow periodically leans up or sits up & looks around as if bored. it sulks & moves slowly, uncomfortably. *if someone touches it on the corners it twitch some *pressure on the center will make it relax & the corners to move happily. it likes to be rested on, but doesnt want to shake off people. *ideally it will breath slightly when laying down, and warm a little if there\'s pressure on it, but probably wont have time for that by thesis presentation | |
| Methodology: | *sketches of pillow to familiarize with what it looks like
*stop motion animations to see what movement looks like *flash animation of target movement of animatronic *video of various people making the pillow emote *It is my desire to make a creature that seems alive, even if it does not seem real. Because of this I want to provide my animatronic with input sensors to trigger animations, rather than having it perform a series of animations. Again, this is because I wish to create a semblance of life, and I believe that would be shattered by seeing a creature react to nothing, to watch it perform for air. This is what the animatronics at Chucky Cheese do, and I believe it breaks any possibility of emotional rapport with the characters. Such so-called animatronics are really robotic puppets, existing merely to do a repetitive task for humans. Another reason for making my animatronic reactive is to see whether or not it will develop its own quirks and personality. While I think authorship is necessary to creating a good animation, and I already have personality based on stimuli and being a displaced pillow developing for this character, I want to see non-rigid programming might give rise to unique personality. For example, if the pillow receives 2 conflicting stimuli, which one will it react to? Or both? What order? Or might it even develop a third, alternative response. It is my hope that this creature will not be just the sum of its parts and code, but rather one of emergent behaviors and quirks. I\'m not claiming that I have any chance of creating artificial intelligence, but I hope that this project can also be an example of, if not free will, than at least the freedom of chance. *Animation has traditionally been restricted to flat, 2 dimensional screens or projections and has developed and evolved throughout various cultures in this realm. I will attempt to transfer the coda of 2D creature animation techniques onto a 3D form. In doing so I hope to learn what gestures remain effective from 2D to 3D in creating the same sense of life that traditionally animated characters can evoke. In the field of Human-Computer Interaction (HCI) or Sociable Robotics, the focus has traditionally been on re-creating facial expressiveness. I wish to focus on body language, posture and gesture as effective ways to convey emotional states and responses. My creature may not have eyes or even a face. The Synthetic Characters Group at MIT create their characters to \"search for the \'computational scaffolding\' or alternatively \'conceptual primitives\' that form the basis of commonsensical behavior and learning.\" The Sociable Robots focuses heavily on facial expressiveness. Pearl @ CMU is designed for assisting the elderly & finding ways to fit into a more dependent lifestyle, like other Japanese Nursebots. | |
| Conclusions: | *what am i DOING?? | |