Great videos, I watched all of them through!
I wish I had something constructive to add here, but all I can say is Awesome!
I think feedback might be minimal because there’s nothing to really compare this to (for a majority of blender users anyway) so people are just hype that it exists.
All looks good to me
The only thing I would want is a bone for the masseter muscle:
Here it is in action, watch Tuvok’s cheeks (brown guy with pointy ears at 1:11):
I understand that this might be difficult (maybe impossible) to capture from front angle video footage. What do you think?
I have been wondering about the same thing Auuman Anubis.
Given the tracking tools we have now is is possible to: Create a track for the front then one for certain markers on the side, use the modifier stack to blend the influence of the front and side markers to an additional empty perhaps in a different layer, and use that new marker to drive a bone?
I think to make that work, each camera angle would need their own tracking marker set-up. The side videos would need the L and R markers to track the same place (overlapping), but the marker influence would only be 100% on the side that corresponds to the video (Left footage 100% influence on L bones, 0% influence on R bones) and then do the same for the Right side video.
if for instance you blend the front and the side markers that are visible from both cameras you could use nearly 100% of the z and the x coordinates from the front camera and nearly 100% of the y from the side.
Depending on the camera angle of course.
I’ve been thinking about this idea to video a walk cycle from the front and side. Putting markers on my waist that can be used to stabilize the footage before tracking.
I’m still not sure how to setup the shoot to get the best results though.
Of course if this can be done than it could be done for facial mocap as well. However I only have one tripod at the moment and I think that each camera will need to be set at the same height before beginning.
Hello Re-Face users! I’ve updated the addon download to contain a PDF of the documentation, since I was having a lot of trouble with Blender Market’s documentation editor. Everyone who has paid for the addon previously may download the updated addon with the documentation PDF included, since you won’t find that information on the Blender Market product page any longer!
Hey, sorry for the (very) late reply; multiple camera support is on my radar, but not something I can tackle myself right now. It’s something I’d be interested in developing to serve a personal animation project, but it’d require obtaining another camera, and modifying my existing DIY mocap helmet rig. I’m also very busy at the moment, so I can’t devote any time to the development, but multiple camera support is only out of reach right now for those reasons. Otherwise, I have some ideas about it, and it’s feasible.
I don’t think Re-Face will be terribly useful for full body mocap; it wasn’t really designed with that in mind, and there are quite a few other tools out there for importing and retargeting full body mocap. However, I wouldn’t try to stop someone from re-purposing the script and finding other interesting ways of putting it to use.
Good morning. This is my first post here and I’ve been heavily researching ways of getting the video tracking data from a HMC, using a two camera setup (I’ve designed my own HMC which can be seen here https://youtu.be/pfMpAHS06nI), and after looking at professional solutions, Blender is now on my radar.
Now my question is, could I use reface to create the joints I need (my face is stationary), solve those joints, and export the animation data to be used in Modo (my primary 3D app). And I need to be able to solve from two cameras, left and right (just some ideas to throw out there).
And the resulting data will be used on a custom facial rig for retargeting. Been watching this tool for YEARS. Thank you.
