Quick poll, which would artists prefer a developer to finish first,
fast automatic retopology (mixed integer quadrangulation)
or improved masking in the compositor for live action footage.
You should make this into an actual poll (so that members can vote on it).
I would vote for the automatic topology.
+1 to skykooler’s comments. Both of them.
Auto topology please…
Retopology for me.
+1 to fast automatic retopology
Personally having a poll will make this much easier. But in any case, I would recommend to have masking tools for the compositor. There should be more focus in this area equally as much as the meshing tools. Just a point of view.
I guess since Blender is primarily a modelling/ animation package that re-topology should come first.
Personally… +1 masking!
Retopology. This is about gsoc, right? Is the quad-dominant remeshing project going to be used as a foundation for retopologizing?
Hi,
I tried doing a poll but when I edited I didn’t have the option to do so.
No this isn’t gsoc related, a dev has volunteered to do both projects but he wanted to know which one was a priority for users, I said was pretty sure the auto retopo would be by a large majority, but figured I’d ask.
Ah someone was kind enough to add the poll, thanks moderator.
I personally don’t use masking for live action footage, but retopology would allow for much better use of the sculpting tools and the future unlimited clay tool because you can get a nice clean mesh based on what you created and thus allow easier texturing and shading work, plus the retopologized meshes can be used easily with external renderers if needed.
So one vote from me for retopology.
Voted but sad 
compositor mask is losing. I guess it is a 3D package afterall.
No problem 
Oh and retopo gets my vote.
Note this is just which would be first/prioritized, not one or the other period 
Voting for masking, because the feature’s long over due and the next OMP(open movie project) will probably need it more.
edit:
seemed pretty clear to me from the first post that this isn’t a “only one gets done” deal,
suppose some who have voted under that assumption might wish to reconsider,
but alas, the vote’s been cast.
Can´t we have both first - or something else?
It´s kind of a tough choise.
To unleash the full power of the retopo, sculpting and mesh editing tools should be improoved first.
I can do sculpting in sculptris sure, and then import to blender for retopo.
I can also use 3dCoat for sculpting, then I don´t need automatic retopo in Blender.
I could also sculpt in blender and retopo with Topogun.
I could also sculpt in sculptris and retopo with Topogun.
… but I feel uncomfortable to do it all in Blender atm.
To unleash the full power of masking, it should have matchmoving, and the internal renderer should be improoved for better results with real footage compositing. Would the masking have matchmoving?
I could also create matchmove mask with some other tool and save alpha channels.
I could also render with an external renderer.
I could also composit with aftereffects.
… but I feel uncomfortable to do it all in Blender atm.
I really love to have new toys in Blender, no doubt, but I would wish for a more complete pipeline within Blender.
So it might be intresting to know what other improvements are planned to fit in, close the gaps. Would make a choise more easy.
Maybe wait until after the GSOC projects get picked? If a retopo student gets in then the compositor work can happen, then after that anything needed in the new GSOC code can be added. Or it could be a mentor/student thing happening.
Hey all, this has nothing to do with GSoC. This was a request by a dev who does occasional contributions just curious on what the communities priorities were. He plans to do both so it isn’t an ‘either/or’ proposition just a which happens first. While if we accept a retopo proposal for GSoC it would be complementary with this work not replace it.