Slashdot Mirror
Cubism For CG And Movies
Aidtopia writes "Computer Graphics pioneer Andrew Glassner has a cool page on virtual cinema. The Matrix Reloaded introduced us to virtual cinema--re-rendering live action to show it in a way that would be difficult or impossible in real life. Glassner takes this much further by using unusual (and physically impossible) camera distortions, morphing multiple points of view simultaneously in single continuous image. Could this be the next big revolution in film? How long until we see a movie done like this?"
You can read the original PDF paper here
--Tim
...to virtual cinema???
Oh, is that why it sucked?
(just kidding, it sucked for entirely different reasons)
I don't see where the submitter gets off claiming that MR introduced us to *any* new cinematic technique, except perhaps for the fight scene with 200 Agent Smiths and not only was that done poorly but the whole thing could have been avoided if only Neo had done another one of his Superman jumps. In other words, it was gratuitous.
Yeah, I'm sure we'll see cubism in movies. It's another knob the show business kids can turn that will make their latest turd appear "original" and "daring", but I bet we won't see intelligent use of it for several years or more, not until a director actually has need for the effect as part of the narrative. Terry Gilliam's Fear and Loathing in Las Vegas might have benefited, for instance.
BTW, that's one of the things that made the original Matrix so unique... it's use of bullet-time was one of the very rare example of a new special effect that is put to intelligent use right off the bat. What a great movie.
Is this truly the only Earth I can live on?
It is looking a little slow already. So in case it goes down, here is a link to the google cache.
"I don't need a compass to tell me which way the wind shines." - Mr. Furious, Mystery Men
When do we get to see a good movie with a good STORYLINE again? The Godfather and the LOTR series are excluded because they are originally written works. I mean, Matrix 2 looked cool yet it was still boring as hell. I don't need to have a degree in temporal mechanics to undersstand it, I need some serious acid instead! People want more story, less bullshit and Alyson Hannigan nude scenes.
Hate me!
Reminds me of Panquake
http://wouter.fov120.com/gfxengine/panquake/
I vote now to construct a counsel of Holy and/or Wise Men who can seal this technology away to prevent Quentin Tarantino from abusing it.
We could then possibly, umm, have Quentin Tarantino sealed away as well...
/* * pope1 */
Video games have used camera morphing and strange distortions for a long time. The Matrix was the first move I can think of that used those techniques successfully. They look cool and were good for a few movies. But taking them to the extreme is always going to feel like a Matrix/Video Game rip-off. Instead of making a movie that uses every excuse for a new morph, how about using traditional cinematography for 99% of the film and using one or two really cool and appropriate morphing effects.
Don't get my wrong, I love the effects. They look great. But c'mon, when someone has a good idea you don't beat it to death. You subtly modify and expand on it to create something unique and equally pleasing. The movie industry seems to lack creativity lately.
...less Alyson Hannigan nude scenes? Bite your tongue!
"A great democracy must be progressive or it will soon cease to be a great democracy." --Theodore Roosevelt
From these "modified" CG applications, how far are we from completely CGI movies that are indestinguishable from real life?
Final Fantasy is the closest we've come, but it was still clearly CG. If you try to, there are a few brief seconds where you can suspend the belief that it's CG and it actually looks real. Maybe in the future it won't take effort... but instead will take effort to see that it is CG instead of live-action.
Would a completely CG movie be economical? Beyond just the "geek appeal" of a pure CG movie, I mean... In mainstream movie making, could CG characters eventually be cheaper than "real" actors? Somehow I doubt it.
Stewey
There are 10 kinds of people in the world. Those who understand binary and those who don't.
It's not there yet. Either the technology or the animators themselves. I hate to beat a dead horse, but the article already brought it up, matrix reloaded felt a lot more like the spirits within than the origonal matrix.
IMHO, MR was ruined by crappy CG. They should have done all the same stuff using bullet time instead and it would have come out a lot better.
I'm not anxious to see the next disappointing CG movie.
As I think was said above, once this technology gets popular it is going to be abused. There are going to "trippy" movies where every scene has twisted backgrounds or characters and it's going to be so much "art." It's just like when someone gets Photoshop for the first time...every single image they produce comes layered with filters. Ever had a friend who was a guitarist and hung around with him when he got a new Wah pedal? Same thing...constant wah effect. It's pretty much human nature to beat new, innovative things to death. The challenge is finding the newest stuff to beat. I guess this is it.
Does your favorite actresses' boobs look a little large in that magazine or tv advert? Her waist a little thinner? It is already a mainstream business practice to make products and people more appealing to audiences. I've even heard they did it in the Charlie's Angel's movie to make Drew Barrymore look thinner.
"God fights on the side with the best artillery." - Napoleon, Marshal of France - speaking truth to power
I've got a convex mirror and a fish-eye lens. Anyone want to fund my startup special effects company?
-- Posting with Karma to burn.
Real simple.
The future of cinema isn't gonna look anything like what this article talks about. It's obvious. Every person i've shown this to has had their chin hit the floor.
Bowie J. Poag
I've been amazed at how much information we have learned to take in at once thanks to TV and computers. Commercials have become very good at hitting us with images that please us and make us identify with a product in a million ways in 30 seconds. Look at the coors light commercials from the "twins" campaign seen so much during football games last year. Amazing.
This technique takes it to a whole new level by throwing so many points of view at us at once. At the moment, we pretty much get information (and an emotional response) on one person or or thing at a time. This is going to let us take multiple people into account all at once. At first, we (as a movie watching culture) will be slightly confused by the images, and the cuts could not be as rapid as in the matrix. But, once we get used to it, we can combine quick moving images with distorted perspective to make people get LOTS of information at once.
Personally, I think it's going to drive us nuts, literally. It would take a lot of work on your brains part to take in all of this information at once. Trying to reconcile what two people are feeling at the same time (imagine the two people's emotions are at odds!) and come up with an appropriate emotional response. I think after a few years of this, a new disorder will pop up that will make ADD look like normal in comparison.
Another item I wanted to bring up was Richard Linklater's movie Waking Life. Aside from being one of my favorite flicks of all time, the film cast aside a traditional narrative structure along with using some really interesting visual techniques to emphasize the experiences of the characters. I'm not knowledgeable enough to accurately describe these techniques, but they involved to a large extent moving perspectives around, showing characters faces and bodies distort themselves depending on what they were feeling or saying, or having objects appear out of nowhere to provide a sort of running commentary on the current scene. I believe the majority of the film was filmed digital and then overlayed with animation 'effects,' for the lack of a more superlative word...effects doesn't approach what this movie is. Check it out if you haven't seen it (and if you like rambling philosophical non-linear films with a lot of visual beauty).
I'm going to take this opportunity to pimp Godel, Escher, Bach: An Eternal Golden Braid, as it discusses how images of distorted and recursive perspective like this reflect the nature of our consciousness and perception of our environment, among many other related topics.
- Warriors of Zu Mountain
would be a good example. Once people get over the newness of all these special effects from Asian film makers, maybe we can go back to go back to action scenes where people actually do THEIR own stunts unassisted. THen we'll never have to see Jackie Chan on wires again. *sigh*Tea and kung-fu. Life is good. Rising Phoenix
Andrew Glassner *had* a cool page on virtual cinema. Then it was slashdotted.
You can mod your friends, you can mod your nose, but you can't mod your friend's nose.
For those of us that took history, we know that a very dominant and well-respected branch of painting before photography was naturalism. Both in landscape, of persons, and otherwise they strived to create perfect copies of what they saw, to "capture" nature in a picture.
Then along came photography, which although the pictures were poor at first they took much of the glory out of it. A simple machine was doing what artists of traditional art schools had had a "monopoly" on for centuries. So, art took lots of new ways and became more of an art of expression, not representations of reality (though there were of course many such artists before that time, too).
The same with CG. Now we strive to reproduce reality, like in Final Fantasy and similar (at least the people, I won't speak for the spirits). But once that is done reasonably well, I have no doubt we'll see movies taking the power of CG turning it to other expressions of art - I hardly think reproducing the actors like a photography does is the pinnalce of CG evolution.
Kjella
Live today, because you never know what tomorrow brings
Seriously, though, it's like everyone's doing it.
Anyone who listened to George Borshukov's talk about virtual cinematography and the Matrix Reloaded at SIGGRAPH 2003 knows how much work went into modeling the characters for all the virtual cinematography scenes. Referring to the first Matrix's bullet time techniques as being "the same" as what's going on here makes no sense, since they're two totally different means of achieving that type of camera style.
Given the stills that were shown during that SIGGRAPH presentation, which, incidentally, had side-by-side comparisons of the real and virtual actors shot under controlled lighting conditions, there's basically one thing that gives the digital doubles away--somehow, I doubt people were checking the shape of the back of Agent Smith's head during the Burly Brawl.
I guess now I chalk a lot of this up to inability to completely suspend belief when I'm watching scenes that I know are physically impossible to shoot.
Also, some of those Burly Brawl shots are just head replacements, so look for those Agent Smiths that don't quite have the right face and then come back and tell me that all those CG bodies look fake.
Cubism, being a movement in painting that attempted to depict a more complete illustration of the painted subject by showing it from a number of different perspectives, was influential in forming the visual depiction of many directors/photographers. Here is an analysis of Truffaut and Eisenstein making the same argument
Surrealism extended cubism into the fantastic world of dreams and provided a fresh perspective that allowed the auteur to look at what did not exist before, recreating reality, as it were.
Surrealism is not new - check out Salvador Dali's own rendition of the dream sequence in Alfred Hitchcock's Spellbound in 1045.Here is a list of some films using Surrealism in some form to render their visions of reality
Roger Corman talks about Surrealism in his films
Here is a good list of surrealism in films
direction, acting or cinematography aren't the key creative process in filmmaking... editing is (and because of the editing techniques he pioneered, D.W. Griffith can be said to be the father of modern cinema). Of course this all depends on if you ascribe to the auteur theory...
An implicit question of the last, oh, 28 years in movies (using Jaws as the start... maybe 2001?) is if big special effects are an editing process (and, thus, creative) or just the next step in set design? Bullet-time and wire-fu may be neat tricks, but do they add anything to what the story is saying (heck, it's quite possible that movie A can use both to say something while movie B can just use them as garnish)? But even The Matrix's big selling point isn't the action (or what differentiates it from say Ballistic: Ecks versus Sever). If it was limited to non-CGI techniques from 30 years ago, would the movie suffered anything more than "realism"?
So this article has this neat cubism thing. Another tool in the workbench. But film isn't painting. Visuals are a means not an end. Maybe someone will come along and blow us away. But Memento and Irreversible work by using a cut and paste method developed a century ago, not an advance in digital postproduction.
What is music when you despise all sound?
This pretty interesting and his example drawings did look like they offered a fascinating view. My only question is if we people could handle it.
Take Fox's "24" for example. I don't know if they were the first, but I saw it there first so please excuse me if credit seems to be going to them.
Every so often, they'll show several frames of different perspectives at once - but each in its own box/frame (like a pictures frame). On occasion, especially when not too much is happening, it's not hard to watch them all. When something really starts happening, they focus on that particular frame and continue. If they were to do that through the whole show, I would think it would be too much to take in and you would miss things.
Now add this cubism approach. The frames are no longer isolated but morphed together. Looking at a single pictures like on his site still takes a little time to determine what exactly you're looking at. Adding motion would most likely complicate that. If that were done through the whole movie, I would think you would most definitely miss things...and would probably need a large bottle of asperin afterward.
Even take peripheral vision. Unless someone is purposefully trying not to look at something off in the "corner of their eye", the observant person will notice something in their peripheral vision and turn to it. This may not be the best example of how people like to focus on things, but it does add to the question...
That question being, like my subjects asks, can we people handle such imagery? What does the /. community think?
How about the "split screen" shots in commercials where one housewife is scrubbing mountains of dirty dishes, while in an identical household the other is leaving a sparkling room because she's used Sudzy brand soap?
Or, more usefully, the picture-in-picture golf sports shots where you see a widescreen of the golf course, and a closeup of the putt in another window?
Or how about when a signer for the deaf is added in a little oval window on the bottom corner of the screen?
How about the instructional guitar videos, where there are three shots - one so you can see the fingering of the chord, the picking pattern, and an arial shot? Plus, there may be music notation composited in as well.
Nothing especially new here, especially since it's filching from cubists. No one even paints in that style anymore - why emulate it on film?
Remember when video could first stagger frames in the futurist style (sort of like mouse trails - think Nude Descending a Staircase). That was overdone to death, and fortunately we never see that effect except on bad sci-fi reruns.
Special effects are best when you don't notice them, and let the story stay in the forefront.
But it all ignores a fundamental neurological truth: the part of your brain that says "that's a cool idea" (or anything else) is a nice one, but it's not the one in charge of figuring out what's going on in a scene. Anyone who has had sight from birth is pretty well hardwired to spatially understand things from a three-dimensional model consistent with our ordinary experience.
As a result, while techniques like this one can be intellectually satisfying, they really don't serve the purpose of narrative -- sure, you're presenting the information in a more efficient (and intriguing) way, but we can't process it nearly as quickly. The film becomes something that has to be mulled over, rewatched and considered to be fully appreciated -- and the gimmicky nature of the technique can only distract from any real emotional resonance that the underlying work has. Such a film is only really going to succeed on an intellectual level, and consequently it's automatically going to be shoved into the "art film" ghetto -- where these techniques have been all along.
This is cool and all, but it's really just a digital polishing of ideas that have been around a long time. I don't think this guy is going to find his voicemailbox full of frantic messages from Jerry Bruckheimer.
You were supposed to be in Hollywood, making sure they didn't screw up my favorite movie.
Not posting the good plot ideas on slashdot 12 months too late.
Nice going bill, you dropped the ball on this one.
Human perception is interesting.
:)
There are reports that the first people who heard a phonograph thought there was an actual orchestra. They found the effect *real* enough, even though it would sound scratchy, fuzzy, and fake to us today. As we become more familiar with a technology, our expectations go up. It is possible to spend $10,000 or more on a stereo system and still complain that it just isn't the same as live music.
We tend to project and fill in the details: finding shapes in clouds, seeing a face on the moon and on Mars. Maybe we start by filling in the details and then get more sophisticated (or lazy) and expect the technology to do more of the work.
On the other hand, realism is more than just "making it look real". You could argue, for example, that The Simpsons is more "real" than Leave it to Beaver. It certainly is a more accurate portrayal of modern attitudes.
I know your post is talking specifically of kind of a "Turing Test" for CG: can you tell which portions are CG and which were not. I am just continually fascinated by the way humans work.
As a humorous aside, it is ironic that we continue to raise our standards for CG in movies while our expectations for human actors seem to be in serious decline
Oh no, it's true! The future of cinema is going to look exactly like the past.
When I saw the filename, I feared it would be exactly what it is.
Oh man, it's like I'm 6 years old again, and forced to hear all my mothers favorite music. No wonder I'm a synthpop geek now.
There's 10 minutes of my life I'll never get back!
I know what your doing, thats a Sciencetologist(sp?) recruitment video!
You can't fool me!
Hear hear. I hope that, lost in the noise of "it sucked" comments, there are more people who rightfully respond with, "no, you just didn't get it."
Were some of the scenes over-long? Yes, the Smith scene and the dancing could have been cut in half. However, it's very disappointing that people focus on these problems and conclude that the movie was a waste of time.
I don't believe critics who say there was no plot development are thinking through the issues presented in Reloaded. Think of it as a blurry picture that comes more into focus: while the subjects haven't moved much, the additional detail can provide much more insight into the situation -- and what might happen next.
Raising the concept of backdoors, keys, and renegade programs illuminates so much of the background, and implies so many repurcussions (some of which the parent mentions), I'm surprised more geeks didn't enjoy the movie for that concept alone.
Suddenly, there's no "one" AI that's controlling the Matrix. And, significantly, the Matrix isn't a single-function program (to keep humanity enslaved). It's more of an operating environment, in which separate AIs with their own (sometimes conflicting, often independent) desires exist. This completely changes the fundamental concept of the Matrix and, if you think about it, exposes many of the Architect's words as half-truths at best, lies at worst.
As the parent says, think in terms of control. If you were writing a program with the kind of importance and autonomy as the Matrix, would you let a "known bug" run around and possibly bring the system (and civilization) down, particularly when most of the "bug's" choices need to be made outside of your control (ie., the "real world")? I think not -- you would put your program in a carefully constructed sandbox, maybe two.
This, again, changes the fundamental assumptions we were given in movie 1. "Reality" isn't reality. It's simply another construct. If not, why would Neo have power "outside" of the matrix? Why would a Smith clone be able to control a human? (Think of the look of surprise on Smith's face when his clone gets sucked into the phone line.) Is the Council Leader the prior "One" (per parent), or a more subtle AI, working to manipulate Neo into making the correct choice? ("Correct" from the AI's perspective, anyway).
Or, think about it this way. The AI knows that certain types of brain (the conspiracy-theorist, the paranoid, the hacker) will always question the Matrix. Rather than lose these "crops", why not create an alternate Matrix, one which feeds their paranoia? By letting them think they've dropped out of the Matrix and are fighting against it, they would happily live their lives thinking they're free--while still under control.
Meanwhile, we're introduced to new allies, new villains, and a clear view that human political maneuvering continues to play a big role in daily life. Seeing Zion, with Neo being the quiet savior while Morpheus acts as the bombastic orator with the cult of personality, made even the dance scene tolerable for me. How these opposing forces work out in the next movie will be very interesting.
That film, Revolutions, is named with the typical ambiguity of the Wachkowskis. One matrix inside another...where does it end? Will there be a revolution, or only another revolution? What is the real real world like? As my friend said, if Revolutions ends and the camera zooms out to show it's all a kid playing a video game ("Now available! Play on Xbox, PS2, GC or PC!"), we're going to hunt down the brothers W and lay down some serious hurt.
No plot? Heh. Watch it again.
That's kinda what I think. It's like Superman. I actually had that conversation a few times with different people. Why is there no really good way to make a guy look like he's flying? The better the special effects get, the bigger a problem this is. If you just go with a cartoon or a bluescreen, you can get away with the fist out in front and the sad little cape-flap, but when you have top-of-the-line CGI, it's way to easy to get bothered by the whole thing subconsiously. The more detail with aerodynamics and G-forces you get right, the weirder the whole "hey! he's flying!" thing gets.
Matrix had that problem. If you think about it, there wasn't really anything wrong with the way they made Neo fly, his coat flaps fine, he has that crazy wake thing going on, but he was flying and dragging cars along in his wake. Whether or not that would really happen if you flew a guy through a city at 90 billion MPH, it looks wrong.
That's what I like about this cubist guy. Maybe not cubism specifically, but we need people thinking about non-realistic ways to shoot things. We need somebody to play around and find out what trips off your subliminal wrong-o-meter and maybe make a Superman animation that isn't completely mathematically correct, but looks way better than one that is.