CPSC 426: Assignment 1 (10%) Due in class, Wednesday, September 30, 2009

1. (5 points) Find and read the article “The Reality of Simulated Actors”, by Alvy Ray Smith, Communications of the ACM, July 2002. Google Scholar is a useful resource for finding papers. You will need to be on the UBC network in order to view the version of the paper found in the ACM Portal. Comment briefly on whether you agree or disagree with the author’s assessment and provide a supporting argument. Discuss any films, games, or other media that have appeared over the seven years since the article appeared and which might pose a challenge to the authors opinion.

2. (5 points) This question is about gaining some brief experience with traditional image- based animation. Use one of the following online animation tools in order to make a simple short animation. Give a URL for your final animation. The best animations will be shown in class.

http://www.aniboom.com/ShapeshifterMain/ http://goanimate.com/ tutorials at: http://goanimate.com/howdoesitwork http://www.fluxtime.com http://www.ratemyanimations.com/

3. (2 points) Develop the basis matrix for a cubic parametric curve where the user specifies the starting position, velocity, and acceleration, as well as the ending position. Do not bother with inverting any matrices; writing the answer in terms of a matrix that needs to be inverted is fine.

4. (7 points) Implement a basic OpenGL program that demonstrates the use of Catmull- Rom splines to control the 3D movement of a ball through a simple scene. The control points should be read in from a text file. The demonstration should alternate between two forms of playback, the first which linearly advances the curve parameter over time, and the second which has the ball move at a true constant velocity, i.e., using arclength retiming. As an alternative to the Catmull-Rom example, you may choose to demonstrate a simple 2D or 3D object undergoing an animated deformatin using free-form deformations. Do not hand in any printed code. Create a README.txt file that includes your name, student number, and login ID, and any information that you would like to pass on the marker. Create a folder called ’a1’ under your cs426 directory and put all the source files, your makefile, and your README.txt file there. Do not use further sub-directories. The assignment should be handed in using: handin cs426 a1

Page 1 of 1 36 July 2002/Vol. 45, No. 7 COMMUNICATIONS OF THE ACM [BYALVY RAY SMITH] The REALITYof SIMULATED ACTORS The acting of actors will survive the digital revolution; their on-screen appearance may not.

s there some elemental quality that is exclusively cult than commonly understood; at the very mini- Ihuman? Must actors portray it for their charac- mum, it will require a Moore’s Law increase in com- ters to be believable? Or can they be replaced puting power of four orders of magnitude, as with digital simulations? Should they therefore fear promised by 20 years. Using the “10 in 5” for- for their jobs? What do actors and animators have mulation—anything good about computers gets 10 in common? How will they interface with their on- times better every five years—we will probably need screen avatars? Can an actor-animator collaboration 10,000 times more computing power within a typ- win an Oscar? ical movie budget than is currently available. Almost two years ago I made my first attempt at Cameos of increasing length will appear in the predicting when, if ever, we would be able to replace meantime, as the techniques of human representa- human actors with digital simulations [2]. Here I tion are mastered and Moore’s Law yields suffi- revisit and refine those predictions. The gist is that ciently cheap cycles. we must separate acting from (the appearance of) actors. I therefore expect that (1) we will not replace Consciousness acting, nor therefore actors, in any known way in Prediction 1 follows from the lack of any successful any known timeframe. But (2) we may well replace theory of consciousness. That is, there is no known the appearance of actors in my lifetime. I have in way of describing how to make a machine, includ- mind a parallel to the goal my colleagues and I once ing ourselves, conscious. We only know that we, at had, starting about 1974, to realize the first com- least, are conscious, so it will probably be possible to pletely digital feature film. That took 20 years, far understand consciousness someday, but that is a longer than originally guessed, and the result, Toy statement of my personal religion, not a scientific Story [/Disney 1994], was a cartoon. I now prediction. The best current theory of consciousness predict it will take an additional 20 years to produce is that of neuroscientist Antonio Damasio [1], but the first completely digital “live-action” motion pic- even he does not attempt to explain a fundamental ture, including by definition fully realized human aspect of the problem, the “qualia problem”—a tiny beings. Explicitly, this will be demonstrated by the example of which is how we derive “blueness” from complete replacement of the appearance of a lead light of certain frequencies exciting the electro- actor in a feature-length motion picture, including chemical system of our retinas and the brain behind comparable amounts of screentime and spoken dia- them. His theory posits that consciousness and logue, and comparable numbers of close-up and emotion are inseparable, and that consciousness is medium shots. Moreover, I predict that the screen in fact a feeling, based on his clinical observations representation, including voice, will be “driven” by that the removal of brain parts causing emotions to at least one accomplished human actor, a member cease also causes the cessation of consciousness. It is of the Screen Actors Guild. impossible to simulate acting if we cannot under- colleagues have questioned my time estimate as being too conservative, but I Dr. Sid and Dr. Aki Ross from Final Fantasy: The Spirits stick to it because solving the problem is more diffi- Within (© 2001 Square Pictures, Inc., Columbia Pictures).

COMMUNICATIONS OF THE ACM July 2002/Vol. 45, No. 7 37 stand consciousness and emotion. Therefore, acting In another five to 10 years we will see 80 megapolys per requires actors; there is no known way around them. So frame as an average frame complexity. But that is only a the remainder of this essay concerns Prediction 2— measure of satisfactory stills. replacing the screen appearance of actors. A successful representation of a human actor must A key point is that animators are actors, though silent move accurately, too. This is where the control problem ones. When I met animator Frank Thomas, one of the confronts us. Woody in had about 100 con- “grand old men” at Disney, several decades ago, he was trols for his face alone. Al (the scheming proprietor of acting into a mirror to inspire his animation of the char- Al’s Toy Barn) in had about 1,000. Each is acter Sir Hiss in Robin Hood [Disney, 1973]. Today clearly a cartoon. It is conceivable that a satisfactory Pixar hires animators based on their acting ability. Ani- human actor’s face might require tens of thousands of mation has always separated acting from the appearance controls. Presented raw to an animator/actor, so many of the actor. We do not tend to think of animators as controls would be overwhelming. The control problem actors because, until now, their screen appearance has involves presenting artists with a sufficiently rich set of always been a cartoon. Their screen appearance—or controls in an intuitive way. I suspect the problem is avatar, to borrow a term from the Internet—has been an quite difficult. I hope to be surprised with solutions that object or a comic drawing of a simple human or animal. bypass the old notions of building a model, driving it Actors are animators. Human actors can be thought directly with animation variables; I have seen several of as animating their own bodies as their screen appear- prototype technologies offering such shortcuts, but they ance, or avatar, and a voice, of course. The really good are not yet perfected. ones convince us that the same body, their own, is that of many different people. They seldom change gender Collaborating with Animators and cannot be animals or objects. A favorite movie, How will traditional actors adapt to this new world? Being John Malkovich [USA Films, 1999], explored the One obvious way is by collaborating with animators, possibilities available when some other actor “drives” the other kind of actor. They already do this in the John Malkovich’s body. In one scene, a woman driving voices for cartoon characters. To be clear, the actors of his body has sex with a woman. Is that hetero- or homo- these characters are their animators, not the highly sexual? By the way, it is revealing to see what people touted voice stars. Yet the voices—absolutely crucial to choose as their Internet avatars: sometimes themselves, the believability of a character—affect the presentations but also animals, objects, and, surprisingly often, a by the animators, who are inspired by the gestures of the human of the opposite sex. voice-actors when creating those of the corresponding Two major problems confront us: animators have to avatars. Human actors will continue to do the voices. be given realistic human models to animate; and actors, Actor-animator collaborations will surely play an freed from their bodies, have to be given effective meth- increasingly important role in the future. ods of driving these models, or avatars. That is, repre- Another possibility is that some animators will cross senting the appearance of reality in a convincing way is the voice boundary and come into their own as they a problem, the “model problem,” and interfacing to drive, or animate, increasingly realistic avatars. It has such a model is a further problem, the “control prob- been suggested that awards be given to animator-avatar lem.” Solutions to both will require major increases in combinations as they are now given to actors driving computational power. their own bodies. One thing is certain: Human actors will not go away Thousands of Processors in that future. c The computation of a single frame of a major digital motion picture is much more computer-intensive than References is commonly understood. Each frame of Toy Story took 1. Damasio, A. The Feeling of What Happens. Harcourt Brace & Co., New York, 1999. 2. Smith, A. Digital humans wait in the wings. Sci. Amer. (Nov. 2000). an average of seven hours to compute, and each frame of Toy Story 2 [Pixar/Disney 1998] four years later took about five. The best digital movies today—clearly and Alvy Ray Smith (alvyray.com) is a digital photographer and president of Ars Longa; he was director of computer graphics research at designedly cartoons—require some of the largest com- , Ltd., co-founder and executive vice president of Pixar, founder putations on Earth, several thousand processors running and president of Altamira Software Corp., and winner of two technical around the clock for a couple of years. Academy Awards. My colleagues and I have long considered 80 million polygons (megapolys) per frame the threshold of “real- ity,” meaning a sufficiently rich approximation that audiences cease to be concerned about its authenticity. © 2002 ACM 0002-0782/02/0700 $5.00

38 July 2002/Vol. 45, No. 7 COMMUNICATIONS OF THE ACM Spider-Man (© 2002 Sony Pictures Entertainment, Columbia Pictures, Marvel Characters, Inc.)

COMMUNICATIONS OF THE ACM July 2002/Vol. 45, No. 7 39