I am experimenting with vector blur. I’ve looked around and found what was on the blender wiki and various threads to be most useful on the subject.
But is using an input of renderlayers the only method possible? Is there no image format that embeds the vector information? It would be nice to be able to make adjustments to motion blur after rendering, as opposed to making sure everything is just right before hitting that render button.
BeBraw: The example file you directed me to uses the renderlayer method. I am looking for a method NOT using renderlayers. Although that was a good read, I hadn’t found that page yet.
Problem with using render layers is that the scene must be rendered when creating the composite, which is what I am trying to work around.
However, I did just have a minor revelation… Is it possible to use two inputs: the image to be motion blurred, and an input of renderlayers. However, only use the vector information from the renderlayers to feed to the vector blur?
Perhaps this is what you were trying to show me. Anyway, I’ll give that a shot and see how it works for me.
Ah, I forgot to mention. I am trying to apply this motion blur to a sequence of frames I have already rendered (which is my reasoning for attempting to use something other than render layers. That way I wouldn’t have to re-render the animation.)
I did just that, I set up a render layer that consisted of the object to be blurred with only ‘vec’ enabled and fed its speed information into the vector blur node.
i think the OpenEXR multilayer image format saves the Vector layer, and is available in 2.43. So, you can render to that image sequence and then be able to use Vector Blur later.