Break images onto cubes face into z-space

[Justin Productions]

Haha! Are you french? Because this guy’s name is Corneille and he’s a great singer in our province (Quebec – Canada).

” We must live every day like it was the last “. That was the message in his song.

I would really suggest you use a very basic 3D application like Blender (not that Blender is not great) because it’s a very simple thing to do.

Doing this in Zaxwerks Invigorator? Hmmm…I don’t know. Maybe John or serge would answer that.

Justin Productions
Tangerin01@hotmail.com
Adobe After Effects 6.0 Professional

[Arnaud Paris]

Thanks Justin, indeed I’m French, Corneille is a great artist. Most of his music videos are quite creative.

I definitely thought about using invigorator with its layer mapping function, but I don’t get how to create the cubes at different depths and have the image sort of “projected” onto them.

Arnaud

[Sebastien Allain]

Hi Arnaud,
It depends on how many mapping you need at the same time.
In fact, invigorator is able to manage only 8 layers… moreover it’s not easy at all to share a map with many different objects.
Another limitation is the number of set… 8 too so you’ll not able to animate more than 8 cubes…

sebastien

[Sebastien Allain]

firstable i have create such effect with expressions to scale layer depending on the distance from camera. Then i use a fractal noise to create rectangular mask on each layer…
Then i have seeing a plug from buena called “depth cue / camera map” which does it by itself. Normaly it is used for camera mapping effect but it works great for such effect.

sebastien

[Steve Freebairn]

I’ve messed around with creating an effect similar to this, the way I did it, was to create a bunch of small square solids that I formed into cubes (for this, they would only need 2 sides) and then I placed 2 light sources (1 behind each “background”) and project that image onto the white solid blocks. If you put enough blocks out there in 3d space, then you’re project image will look perfect when you are looking dead at them, when you start rotating your camera to it’s second position, then your background will show that it is actually a bunch of blocks. Then you’d just place your keyed footage of your singer on top of the rest of that. Hope that helps

[Arnaud Paris]

Sebastien, Steve,
Thanks for the help.

I look into the Buena Plug and it makes perfect sense using their projector camera technique.
Except that I don’t see how to create this cubes in such a way that they fill the screen with randomness in the z axis without having black areas between the layers (or cubes).

I looked into the digital anarchy layer assistant plugin which is good at filling the screen with many layers on the XY axis but as soon as I add some randomness in the z axis they don’t cover the entire screen anymore.

Arnaud

[Sebastien Allain]

I had prepared many layers in photoshop: i decided where each square mask were… not random at all!

Arn”oho” is a japanese inspiration name ? o namae no nihongo desu ka ?

sebastien

[Arnaud Paris]

I see Sebastien, I just thought that there was probably a way to automate the process.
As for my name no voluntary relation to japanese, all the variations or Arno were taken
by other cow members… there’s got a be a lot of Arnos in mograph!!!

Steve how did you manage to get these semi cubes to be at different depth and yet to create
a full image for both of their sides? Doing it manually seems to me like a real brainbreaker, yet you
said it took you 10 minutes.

Here is a link to the project file of what I was able to achieve so far, maybe I’m doing something completely swrong but just that took me the entire afternoon to figure out.
(You need to have the buena depth cue plugin installed)
https://www.sysmicfilms.com/cow/Depth_Cue_Test.aep

A.

[Steve Freebairn]

Ok, I started typing this thinking 10×10 blocks, but that isn’t going to be right, so use 20×20 blocks (so the numbers will be off, but I’m sure you can follow the math)

As far as I could see from reading my post (but I could be blind) I never said it took 10 minutes. but, it isn’t quite as hard as you may think. If you use 10×10 “blocks” you only need 3888 blocks! 🙂 That is a ton if you are going to hand place them, but if I were you, I’d create (and this is for a 720×480 project) 1 solid that was 10×10. Then I’d move the anchor point to the right edge of the layer(you’ll see the wisdom in this later) then duplicate it once. move the 1st instance of the layer to the upper left of the comp window (you shouldn’t have a camera or 3d layers yet) then duplicate it until you have 72 layers. then use align and distribute to place them side by side. Then Highlight all of your layers and duplicate them, before you deselect the layers I’d do a few things, first I’d change the layer color to something else, then I’d change the position of them so that they are at the bottom of the screen. Then I’d duplicate them until you have 3888 layers. This next part is the more time consuming one. You need to select the layers so that you have all of the layers that are one the far left and then use align and distribute so that they form into a column and then repeat that on each set of blocks. This will fill your comp window. Now check the 3d box for all the layers. then you need to duplicate all the layers and Rotate the duplicate of each layer 90degrees and then you need to parent each duplicate to the layer it was duplicated from. Then you’ll just need to slide these layers into z space at different depths (you could use an excell file to make a table with all the positions you need to cover and then randomly move each layer set into place.

And after typing all of that, I think there must be an easier way. I’m sure someone could write a script or an expression that would do this all in a few seconds. Like I said in my first post, I was doing something similar. My similar idea was a lot easier because I just used the wiggler to have my boxes move randomly and then i added enough boxes until it worked out. Sorry I don’t know an easier way.

[Sebastien Allain]

Arnhaut,
Here is the link to show you an old similiar job. As i said, this one was created only with expressions… no buena plugin and no projection throught light: only layers with mask and expressions.

http://www.apt-a.com/cow/temp.mpg

sebastien