Could you add webcam faceposing?
I mean, i make a face with the cam on and the games makes it on a Ragdoll that i've previously selected
Do you have any idea how hard that would be to program?
[editline]3rd July 2011[/editline]
Hell, I don't even know where to start besides: [b]It's impossible.[/b]
[editline]3rd July 2011[/editline]
I mean first of all, the ragdolls don't have enough facial flexes to match your mouth eye and nose movements on a webcam. Say you'd want Heavy to make a funny face, it'd be impossible because the ragdolls can't stretch their faces as much as you can. You'd have to get new ragdolls with hundreds more flexes to make advanced faces like that.
[QUOTE=Dalek;30868317]Do you have any idea how hard that would be to program?
[editline]3rd July 2011[/editline]
Hell, I don't even know where to start besides: [b]It's impossible.[/b]
[editline]3rd July 2011[/editline]
I mean first of all, the ragdolls don't have enough facial flexes to match your mouth eye and nose movements on a webcam. Say you'd want Heavy to make a funny face, it'd be impossible because the ragdolls can't stretch their faces as much as you can. You'd have to get new ragdolls with hundreds more flexes to make advanced faces like that.[/QUOTE]
You have to be careful with the term "impossible", its pretty much impossible but the work required to achieve this is probably not worth itself.
The best way you could do it is mapping certain facial coords to certain ragdoll bones and probably do it like that. Facial flexes per see wouldn't work in this sense.
Which in turn would require somewhat different facial setups to the current system of course.
I don't really see it happening
Face recognition code + face translating code + defining what each parameter inside the models do + output code... I suppose this is needed to be coded.
nope
Really?, they used this for L.A Noire, it is possible, but not anytime soon I suppose
[QUOTE=Dog;30888560][media]http://www.youtube.com/watch?v=Phcdp7AjhBg&feature=channel_video_title[/media][/QUOTE]
And there you have it.
[QUOTE=Zupinata;30885161]Really?, they used this for L.A Noire, it is possible, but not anytime soon I suppose[/QUOTE]
You must have missed the part where that was done with a external application to create pre-baked animations (like basically any animation in a video game). None of it is done real time, and it required a lot of hardware to pull off, you cannot just use a webcam or small recording device to do this kind of thing, you need an entire array of sensors, a good quality camera and perfect lighting.
Doing it in real time in a game also increases the hardware requirements a shit ton.
[editline]4th July 2011[/editline]
And according to that video above you can do it in Source :v: But still, it requires a lot of effort for minimal gain and with what looked like fairly low quality faceposing due to innacuracies.
Sorry, you need to Log In to post a reply to this thread.
