I prefer to  say faceshift is better FOR MY SYSTEM CONFIGURATION, may be faceware is good but is not adapted to my system. I don't know if you'll see the difference but I do no more need to spend hours for correction post mocap with face live. At 0;33 you'll notice a lip movement skipped by faceware, it happens manytimes and I have to spend minutes to correct this




     

Charly - yes, I saw the 'missing' lip response at 33 sec. - but what *really* stands out to me in this test is the "responsiveness" of the eyes (with faceshift) - especially the *pupil movement* .....  Brilliant test mate! ...

--------------------------------------------------------------------------------------------------------------------------------------------------------

Please be patient with me ..... I don't always 'get it' the first time 'round - not even the 2nd time! :(  - yikes! ... 

● MSI GT72VR Laptop, i7 7700HQ 4-Core 3.8 GHz 16GB RAM; Nvidia 1070, 8GB Vram ● iClone-7.93 ● 3DXChange Pipeline 7.81 ● CC-3 Pipeline 3.44 ● Live Face ● HeadShot ● Brekel Pro-Body ● Popcorn FX ● iRAY ● Kinect V2 ● DaVinci Resolve17 ● Mixcraft 8.1

Yes, and when I close my mouth between sentences, with faceshift, lips are really closed and follow the movements. I suspect the difference due to my system, I have 16Go of RAM and only 4Go for graphic memory (GTX 1050)


     

Yeah .... I see what you mean. The lips definately seem more responsive overall. Funny that, because with Mike's tests I *thought* the accuracy was a bit lacking. But tests vary - and *both* yours and Mike's results gave me the impression of *more natural*. But again, unless you see the *live* performance side by side with the puppeteering - you can't assess the *accuracy* .....

Edit: Thinking more about this ..... there's a difference between "responsive" and "accurate" ...
● Responsive ..... meaning the face readily creates *movement* and doesn't *miss* cues.
● Accurate .....meaning (with lips), that the 'phonemes' shape themselves properly in accord with the words spoken.
(This is just my *take* on things).

--------------------------------------------------------------------------------------------------------------------------------------------------------

Please be patient with me ..... I don't always 'get it' the first time 'round - not even the 2nd time! :(  - yikes! ... 

● MSI GT72VR Laptop, i7 7700HQ 4-Core 3.8 GHz 16GB RAM; Nvidia 1070, 8GB Vram ● iClone-7.93 ● 3DXChange Pipeline 7.81 ● CC-3 Pipeline 3.44 ● Live Face ● HeadShot ● Brekel Pro-Body ● Popcorn FX ● iRAY ● Kinect V2 ● DaVinci Resolve17 ● Mixcraft 8.1

ok sonic, failing to give you a iphoneX, I owe you a test side by side even if I don't like to show my"face" , I'll try to do it for you if I have a time :-)


     

Well that's very kind of you charly !!
I'm sure that there'd be *many* an interested viewer regarding this sort of comparison.
You don't 'owe' me :)  - if you get the time *great* -- if not -- no problems, - you've already made some useful tests available :)
(which is a great asset to this forum btw).


--------------------------------------------------------------------------------------------------------------------------------------------------------

Please be patient with me ..... I don't always 'get it' the first time 'round - not even the 2nd time! :(  - yikes! ... 

● MSI GT72VR Laptop, i7 7700HQ 4-Core 3.8 GHz 16GB RAM; Nvidia 1070, 8GB Vram ● iClone-7.93 ● 3DXChange Pipeline 7.81 ● CC-3 Pipeline 3.44 ● Live Face ● HeadShot ● Brekel Pro-Body ● Popcorn FX ● iRAY ● Kinect V2 ● DaVinci Resolve17 ● Mixcraft 8.1

""" WHAT ARE YOU TALKING ABOUT !??? """
Well, faceshift are better result , but I hate any Apple's Phone at all ( I'm Android hardcore ) ...
Hope RL cross-operate with Facerig who's using INTEL face recognize technology ( rumos say is more cheaper ).
Below link is action on this app:

About your "responsive" versus "accurate" assessment:

I've been doing lip sync in various forms for over three decades now, and while I am not saying I'm "expert" I have developed a lot of knowledge about it.  One thing you need to be aware of is that we, as humans, tend to perceive things which aren't there.  If you want some excellent examples of this, watch something that is *really* well dubbed (you can find them out there -- in all forms, whether anime, or live action foreign films, or whatever).  Here's a really good (and fun) one: the classic movie "Goldfinger" had the lead villain played by Gert Frobe, and he could barely speak English, so his entire part is dubbed.  I defy almost anyone to tell.

So what does that mean?  The lips (and phonemes) are obviously not matched up with the sounds, yet it looks convincing to our eyes.  Why does this work and, conversely, there are times we feel the lips are not right even when they are?  It is a VERY complex issue, but it boils down to the fact that we really just aren't that good when it comes to reading lips (this is also why there is a lot of fallacy about "lip readers" -- they will tell you that the best they can do is "come close" and it's far better if they know the context of the conversation, or even hear some sound, because SO many words have exactly the same mouth shapes, and yet in movies they are also so completely accurate).

All of this is a long-winded way of saying if it looks right it is right, no matter how it really happens.  Charly says he exaggerates his speaking for the lips to seem right to him, and that's fine.  Some people, though, speak with lips practically closed and that's fine as well.  I think people looking at these examples spend FAR too much time nit-picking and not understanding that in the context of a scene (where you will NOT be focused on the lips) you won't even notice, and "good enough" will be plenty good enough.

---

Alienware Aurora R16, Win 11, i9-149000KF, 3.20GHz CPU, 64GB RAM, RTX 4090 (24GB), Samsung 870 Pro 8TB, Gen3 MVNe M-2 SSD, 4TBx2, 39" Alienware Widescreen Monitor
Mike "ex-genius" Kelley

Yeah interesting thoughts there Mike. Things I hadn't thought of actually. It makes sense, as you say, if it *looks* right - then it is.
It's true we do tend to scrutinize this stuff. Whether it's faceshift or faceware that's used in a production - there simply isn't quantum leaps of difference between them .....  :)

Hey akuei2 - that's a funny clip lol ... :)  (I'm Android too ;) atm)

--------------------------------------------------------------------------------------------------------------------------------------------------------

Please be patient with me ..... I don't always 'get it' the first time 'round - not even the 2nd time! :(  - yikes! ... 

● MSI GT72VR Laptop, i7 7700HQ 4-Core 3.8 GHz 16GB RAM; Nvidia 1070, 8GB Vram ● iClone-7.93 ● 3DXChange Pipeline 7.81 ● CC-3 Pipeline 3.44 ● Live Face ● HeadShot ● Brekel Pro-Body ● Popcorn FX ● iRAY ● Kinect V2 ● DaVinci Resolve17 ● Mixcraft 8.1

That's what's struck me too. The responsiveness of the eyes movements. In this test, of course the Faceshift take is much better than the Faceware take. But it also depends on HOW you capture the subject. I don't own Faceshift, only Faceware, so I cannot judge Faceshift. All I know is there are settings in Faceware to make the eyes (and lips) much more responsive while you capture. (see pic).






My latest movie THE GOLDEN MAN, and a few others: https://www.youtube.com/channel/UCIl1EqVCKitZzLqaNnLK0BA