Hello everybody!

I have this 3D scan straight out of Capturing Reality. Already cleaned, but with horrible topology and messed up UVs.
I worked with this tutorial and as proof of concept I got a working model with a normal map into Unity.
Now because of a dirty unwrap I have to do it again, and before I start I wanted to check back with you guys, to make sure I do it the mose proper and efficient way

1) Import Tool and Texture
2) Duplicate, create Dynamesh with desired resolution for final model
3) Smooth dynamesh

Is the Idea of this step to loose the artifacts? Because I cleaned my model before importing.
Can I ignore this step if I project my geometry later onto this dynamesh, or is there a benefit in smoothing through the dynamesh?

4) Remesh with guides to optimize topology flow in face etc.
5) Unwrap target

I used a control texture, and zBrush cut through my original Topography. Is there a way to avoid this, or do I have to export, unwrap and reimport? I would prefere to do the unwrap in the End on my low poly output.

5) Polypaint

How exactly does this work? Do my Vertices on the Model sample the texture? Live Vertex Color?
Do I have to subdivide the source and my dynamesh target to achieve a better texture resolution when I project geometry and polypaint? I bumped my source 60k model up to 11Mio, and my dynameshed target to 8Mio to get an acceptable color information.

6) Project Geometry and Polypaint onto Dynamesh
7) Decimate with decimation master.

I want to keep more details on the head. Is there a way to draw two masks that keep a different amount of detail? Now I reduced the model, exported, imported, masked the head, reduced the rest, exported.

Would be happy to learn a few tricks. Very interesting software.
Regards, Achim