Cineversity

I’m trying to “map” the face morphs from Face Moves to a character created by iClone. Like the other iPad app “FaceCap” the morphs are not the same. Though in FaceCAP, I can clear all the imported morphs attached to the captured face and adjust the morph strengths to see what face and mouth expression each morph creates. I can’t figure out how to do the same with Maxon Face Moves as they seemed to locked inside the tag. Maxon Face Moves tag has a property window that shows morph strengths, but when I change a value the face mesh doesn’t change (I think it’s only for reading).

Is there a way to do this or does anyone have a pictorial of each morph?

Hi brucek5,

I’m not familiar with the two apps you mentioned.

The provided slider/parameter animations from MovesByMaxon, will not provide the facial-expressions to a 3rd party mesh. Each pose needs to be defined newly.

If you have many blend-shapes that work for one and the same mesh, those can be dragged into the PoseMorph, while the PoseMorph itself has the same base mesh.

I can provide an upload link.

In case you like to share anything, please use the attachment option here (needs to be zipped and below 4MB), or use one of the following could service:  DropBox, Amazon< Adobe, Apple, Google, Wetransfer. Please avoid shortened or the highly hackable and legendary unsafe bit.ly stuff. etc. Thanks. Images itself are, most of the time, useless. Please just the needed content, no 3rd party stuff, thanks

I have, however, explored some ideas.
https://www.cineversity.com/forums/viewthread/3961/

To my knowledge, there is no reference for expressions available, and I believe most artists would ignore them anyway.

All the best

Greetings,

So first, I have looked at the links you provided.

Reallusion has software “Character Creator”, “Head-Shot” and “iClone” that together you can create a fully rigged and faced morphed 3D character model. “Head-Shot” creates the head mesh and texture maps based on AI using a a hi-res photo. I can than export in FBX format with the built-in settings for Cinema-4D with morph tags and all ready for animation.  The attached image shows a character I made of myself imported into C4D and rendered with Redshift.

The image also shows a face capture FBX from an iPad app called FaceCAP that works much like Maxon Moves. I used it in image because I’m try to explain what I’m trying to do with Maxon Moves.

Since there’s seems to be no standard on face morphs, I’m using Xpresso to try to “re-target” the morphs. You can see in the image I’m adjust the strength of the face capture to see what it actually morphs, then in the next image I look for a morph in my character morphs that produce the same ( or least close) the same morph. You can see the morph names are different and don’t even have the same kind of morphs. Then I adjust my Xpresso with the goal of getting a match.

Maxon Moves also has it’s own set of morphs with it’s own set or words describing them. What I lacking with Maxon Moves is a way of see what each of it’s morphs even do.

I’m surprises I haven’t found anyone else trying to do this as this workflow can create a realistic human looking character and all the pose morphing work is automatically done for you. Just hook-up motion capture and you’re done.

Hi brucek5,

You can just go from Moves By Maxon to any PoseMorph you have it set up. The work is to match those to set the “wires” in Xpresso. Which is kind of a very boring job, to do it over and over again. For that, you like to just build your own translation/connection.

Take the Character Definition Tag. It takes joint information from one animation to another. This creates a “what matches – gets it” connection.
The idea: connect the slider values to a particular joint rotation. This can be used as a standard. In this way, a single capture can be transferred to all facial animation with such an interface realized in your studio. Quick and easy after the setup is done. This would be step one.

Here is an example
https://www.amazon.com/clouddrive/share/29c3U2wrpSe659PoFN8zrRnVNNc3qLtP1VAgXOTamiU

Step two would be to have this “capture transfer” provided from MovesByMaxon. So, every time you need to feed in a new capture to any character, it will be a few clicks and done. Since the poses are defined in MbM, it should be a reasonable amount of work to set this up.

You might suggest that to MAXON directly here:
https://www.maxon.net/en/support-center

I leave that here, so anyone likes to have such connectivity can make his/her input.

All the best