On Floating Windows

Last time, I posted an article about a Creative COW article written by Brian Gardner on the 3D process. In it, he touched on something I didn't recognize that he claims to have invented: floating windows. The topic fascinated me, which may be showing some of my inexperience in the field of 3D film making. I've been unable to experiment with creating them myself yet, so take what I say with a grain of salt, but I've since been watching content with 3D windows, and I think I can explain what they are, why they're used, and why they work.

Basically, the problem they're meant to solve is 3D window violations. This is when a portion of a scene has an object that pops partially outside the screen. This will cause half of an image to be floating there outside the screen, ruining the immersion of the image by calling attention to the fact that 3D is an illusion (the first 3D film I ever saw was Clash of the Titans, which had this problem frequently, initially turning me off of 3D in general). Basically, this will happen any time that an object is partially obscured by the screen.

But Brian Gardener once mistakenly discovered that in 3D cinema, the screen doesn't necessarily need to be fixated in one place. So what if we used 3D cinema to actually move the visible theater screen outward, overcoming window violations because things popping out of the screen aren't popping out of this digital screen anymore?

Apparently, you can do this by cropping the edge of the screen in one frame (either by adding black triangles to a corner, or by adding a black line down the side of a screen). 3D creates depth by creating a physical distance separation between an image sent to the left eye, and one sent to the right eye. By adding a triangle to the corner of the screen in only one frame, or by cropping away a portion of it, the screen itself seems to float out into the audience as well (because the corner is now separated by as much . If you separate the edge of the screen as much as the window violation pops out, then it'll seem like the window violation fits nicely inside the screen. And according to Brian Gardener, if you're changing this effect throughout the film, nobody will notice that it's there.

This seems like a very effective way to fix poorly framed footage, or scenes that couldn't be framed differently due to physical limitations (e.g. a documentary, where you can't move the objects in the scene). It also seems like a useful tool in the tool-belt of someone converting a 2D film to 3D (ala Finding Nemo, Star Wars ep. 1, Titanic, where 3D considerations couldn't be made for the film beforehand).

Anyway, that's what I've been learning about 3D imaging recently. I still need to actually use this technique in my own 3D editing in order to understand all the effects here. It seems like it would be a downgrade, since it's literally removing image information. Although, that's not necessarily always a bad thing. But it seems like something that needs to be understood and considered appropriately for that aspect when you're going to use it, and just frame things for 3D in other situations. Also, I'm not sure how to end this post, since I'm mostly just pondering out loud at this point, but I just did it.

Brain Gardner Is Doing It Right

I recently read an article by Brian Gardener for Creative Cow Magazine. It's an amazing article that does a lot to explain some of the things I've been unable to adequately explain in previous posts on this blog. But more than just establishing himself as a very competent authority on 3D film-making, the article also brought up a very interesting concept about dynamic floating windows. I'm not entirely sure what he means, so I intend to watch more of his work and pay attention for this concept.

What's particularly interesting to me is his filmography includes two movies thought to be masterpieces of stereoscopic cinema: Coraline and Life of Pi. The article references a live action film in the planning stages that utilizes a technique he pioneered, and given the recent release of Life of Pi and how long the article has been up, I think that may be referencing Life of Pi, which so successful that even 3D critic Roger Ebert praised the use of 3D in it. Having not personally seen Life of Pi yet (and I may have to wait until the Bluray is released to see it, because it's no longer playing in 3D in my area), I am indeed very curious of how important that technique is.

What he describes is something he calls a "Dynamic Floating Window," which because I haven't seen it is something of a mystery to me. He talks about a short film called The Black Swan that tried to keep the audience separate from the onscreen action by creating a floating black screen, effectively creating a new theater screen in front of the actual screen. In that way, it would fix something called a "window violation," where only a portion of an object protrudes from the screen. The problem was audiences noticed this new screen and found it ridiculously distracting, causing more problems than it solved.

Brian Gardener says he tried using this idea with a moving screen instead of a flat one, and suddenly nobody noticed the floating screen anymore. It was still solving problems, but it wasn't causing them anymore. To be honest, it sounds too good to be true, so I'm very curious if I'm just misunderstanding what he's describing, or if what he's describing really accomplishes all he says it does without being distracting. Based on the reviews given to the 3D in his work, it's promising.