Stabilizer as a Node?

Is there any way to use the stabilizer feature as a node on the graph? I want to apply a Dead Pixel Fixer effect to my footage but once I use stabilize the footage in tandem with that effect, the dead pixel is now floating around the screen and I have to keyframe the Dead Pixel Fixer to follow it.

If I could node the stabilizer AFTER the dead pixel fixer node, in theory it would be stabilizing that effect with it.

Or is there a better way to handle this? Another user mentioned to fix the pixel, export and then import back into Davinci. However, I would like to work only within Davinci without having the extra workflow of exporting clips and re-importing.

You could try and create a compound clip in the timeline first and use the dead pixel filter in there and then on the compound clip itself you apply the stabilizer after that. Does this make sense?

You could try and create a compound clip in the timeline first and use the dead pixel filter in there and then on the compound clip itself you apply the stabilizer after that. Does this make sense?

You're an absolute wizard! That worked!

Thank you for the insight.

That's totally odd, I never realized that the stabilization is not attached to a node. Am I missing something or is that kind of bizarre?

I suppose the internal logic here is the same as the sizing controls. Those are also detached from the node graph and work on a clip basis. Probably because it is not a color decision when you stabilize a clip.

@TyBailey: Glad I could help

[Peter Chamberlain]

Stabilizing has always been on the source clip before other processing.

Thanks Peter!

Not a big deal, but since you guys are working on the manual, it might be nice to have information reflected in this chart on page 887.

That would be a good idea to indicate exactly where stabilization occurs.

You could try and create a compound clip in the timeline first and use the dead pixel filter in there and then on the compound clip itself you apply the stabilizer after that. Does this make sense?

Thanks for the workaround!!!!!

That said, I could imagine it would be a good thing if there was the option for more flexibility in the stacking order of stabilization without having to create a bunch of compound clips.

For example in my use case, I am applying neat video and a deinterlace filter on some older footage. Having the input sizing and stabilizer functionalities locked underneath those nodes with no alternatives otherwise hinders the effects of those filters.

The middle of the diagram, the operations that take place within nodes, this must be the order of operations within each single node, right?

I would also love if this could be made into a more detailed or updated diagram as I was struggling a bit using the stabilizer together with external mattes on my last project. It seemed to me that in this case the external alpha mattes were "merged" into the whole other processing *before* the stabilizer (had to find this out pretty late in the project and thus it caused me 1-2 allnighters) ...
Btw. it would be great if BM would take the time to maybe put 2-3 more detailed control options into the "external matte" nodes to affect something like this .. as they were also sometimes totally misaligned, then after unchecking and re-checking "lock matte" it "snapped back into place" so it seemed wonky at best!
Thanks

It is also possible to stabilize footage using a tracker within Fusion; this is a different stabilizer than the standard Resolve one, but depending on the footage it can still work well.