How to Render "Volume Scatter Cube" in a 2nd png sequence?

(mikey9871) #1

Hello all,

I apologize if this is easily answered but I can’t seem to find what I need right now.

I’m currently rendering a plain room for testing purposes. Around the entire room I’ve setup a cube with material shader set to a mix of “Volume Scatter” and “Volume Absorption”, which I believe seems to help add a sense of atmosphere to the scene.

So now I’d like to add a human actor (green screen keyed) into this virtual room, but I’d like to make him look as if the atmosphere is also on top of him, not simply him pasted on top of the scene.

How would I go about doing this? I’m assuming I could somehow move the volume scatter cube to a different “render layer”, such that the room is one png sequence rendered, then the atmosphere is a second png sequence rendered. Is that an appropriate solution? Would separating the two elements somehow mess up the lighting?

Thank you for your advice!

(zeauro) #2

If room is on one layer and volume in another one ; you can create one renderlayer for each with same lighting using both layers.

Lighting is mess-up if renderlayer settings are incorrect.

But if both renderlayers are using same scene layers, their lighting should be the same.


Version 1: Place your keyed Image Plane in the scene where you need it.
Version 2: To do it more realsitic with much more control I would do two renderings. Same as above but one with the room and one with only Your keyed human and environment set to shadow catcher and some objects as mask. All the things you need than is composit everything together. Important for this is the z-depht pass and mist for both renderings. That way you have much more control for your keyed human… better matching to the scene and so on. I´m not sure about your volumetrics are rendered in z-depht and mist… But I think it should work.

(mikey9871) #4

I haven’t quite figured out how to use layers yet, but I will try to figure this out. What type of “incorrect settings” should I look to avoid in this case?

(mikey9871) #5

I would like to learn both methods. I didn’t realize I could place a keyed image directly into a 3D scene… that’s cool!

So for Version2, I would put “all things” in one layer, then put the environment+human into a second layer. How would I set the env+human layer to become a shadow catcher, and what do you mean by mask?

Sorry, I’m still a total n00b. It’s amazing I’ve even made it this far, honestly.

(Jason van Gumster) #6

Moved from “General Forums > Blender and CG Discussions” to “Support > Compositing and Post Processing”



Version 2:

is to use the shadow catcher version of blender. ( It is not in official blender at the moment).
With Mask I meant you do the same rendering but in shadow catcher version. Use the overlapping geometries as mask (maybe as a object or material id pass) and the shadow recieving geometry as shadow catcher. Sorry if thats a bit difficult, i believe it should work fine when you need more control in compositing. (maybe you realy dont need for your scene.. i have no idea how it looks like). That way you will have more control in compositing to manipulate your desired result. Thats the way I maybe would do it without the differnet render layers.. but it is somekind of doing the same same like version 1... Sorry.. thats a bit complex to describe. Hopefully you can figure out the compositing and mask thing. It is all about renderpasses and compositing. Good luck! and sorry.. cant describe it better.

(mikey9871) #8

It’s okay Piet! Thank you very much for your explanation. I think some of it sunk into my brain and maybe I’ll be able to give it a good try soon. I’m running back and forth to different concepts and hurdles right now and will bookmark this information and some tutorials if I can find any, based on what you’ve said. Maybe I’ll be back here in a few months asking the same question haha.


He mikey.

maybe hook up to that… It always depends on what you need or want to do. But Im sure you are on the way ;).