Syriana, how did they do the explosion??

XiaoSu Han · February 4, 2009

why wouldn't you need a moco for 3 passes of a dolly shot?

What's interesting about analysing footage like this is that different people see different things. You saw the significance of the man crossing frame. I missed that. You saw it as hand held, but I see it as a track with wobble added - maybe because I've done that myself. What I felt certain I also saw was that the camera was entirely still during the explosion. So a closer look:

The original footage at a better quality:

And partially stablised:

I didn't spend a lot of time on it so a few bumps remain but it's enough. (BTW does this YT Video play ok for you? I've added a huge amount of black at the end to test something.)

It is indeed a track. Importantly, the camera is stationary at the crucial point. After the explosion, when the film's image pans off to the right of the YT window, you can briefly see that the explosion was a static shot and moved left off-screen in post to match the foreground pan. And (or is it my imagination?) the background zooms in slightly as the explosion happens.

I thought I saw some tilt, but stabilising in the x and y axes reveals there's none at all. Interesting illusion that.

No moco needed for this. I think it's very clever and a cheap way to do it. <_<

(The eagle-eyed might notice that all number plates are blurred out. :huh: )

Karel Bata · February 5, 2009

Here's my thinking. The dolly stops before the explosion. After that one man crosses shot masking a change in the background behind the left of Clooney. The amount of pan till then is minimal and reproducible. It only has to be roughly so because motion tracking can produce an exact match in post. The explosion background shot is a locked off camera lined up to match using video playback. Clooney gets closer and vertically divides the frame before the camera pans with him. Everything to the left behind him then becomes a blur. At some point what's to his left changes from the explosion shot to the Clooney shot, but since it's a blur we can't see where.

Now I'm a bit thrown by this account of three shots: Clooney; explosion; and one for the extras. So they choreographed all the extras to do their stuff in one shot? There's a lot going on there to get right. And the four guys crossing frame r to l are problematic to my theory. Getting a good match without moco would be difficult...

Unless those four guys (and our foreground guy) did their stuff during the Clooney take! Then you wouldn't need to get a matching camera move for them.

So why do you need moco? They must have shot Clooney before they blew up the car right? What you suggest is that the shot was planned so tightly that they confidently panned with Clooney knowing that when they later blew the car the timing would fit perfectly.

But hell, the more I look at it the more I'm not so sure. So, how do we find out..?

Scott Fritzshall · February 5, 2009

why wouldn't you need a moco for 3 passes of a dolly shot?

If you're shooting multiple passes for a shot that has a relatively simple camera move, and for which the elements don't depend too precisely on timing, you can sometimes choose not to shoot moco, with the tradeoff that you have a potentially much more complex compositing task. I'll try to describe it for a generic setup- just a take with an actor, and a take with a background. Shoot both as similarly as possible; choreograph it to get the timing right. Do a really precise 3d camera track of both the actor plate and the background plate. Create basic scene geometry representing the contents of the background plate, and project the background plate onto this geometry. Then render this from the perspective of the tracked camera from the actor's plate. Projection mapping in this scenario is a powerful tool that allows you to "rephotograph" your plate from a somewhat different angle, in 3d. Done properly, this removes any differences in camera moves between the two takes.

Here's a half-decent example of projection mapping on youtube:

This one's not bad either:

It basically allows you to change the perspective on a flat image, so as long as your takes are similar enough (you may have to retime one or the other).

Is that helpful?

Karel Bata · February 5, 2009

I don't think you need that 3D element here, only 2D techniques. The 'camera shake' masks any perspective irregularities. Like it's clear they didn't bother with any perspective change when panning the explosion off-stage as would have happened in a real pan. Hell, they even zoomed it!

BTW if you add &fmt=18 to a YouTube url you get the HQ version - even if if you don't see the HQ prompt! So your last link (which is awsome! as they say in net-speak) becomes

What I'd like to know is how some YT users get their videos to fefault to HQ?