dueyftw wrote:
Has anyone tried this software? Is it as easy as they make it out to be? And where do I get a mount to point back at my face. A WiFi camera would be cool.
Also could the rgb of a knect work? :)
I dabbled with Zign Track (version 1) a few years ago. At the time I thought it was way too much work for the results I got--I got more credible results manually matching video reference and that took just a little longer. To be fair though, that was a long time ago, and the software may have improved a bit.
I don't know anything about Maskarad but output results in their screenshots looks similar. However, the fact that it's a markerless system is very interesting. I don't know...I just may have to play with the trials when I have time.
Sorry, I know that wasn't helpful at all but FWIW, I'm intrigued. Probably not enough to spend that much money though. :p
Regarding head mount camera rig: I think it will be difficult to find a rig that will lock the camera steadily enough on your face to be useful, and if you did find one it will probably be bulky and affect the quality of the markerless body capturing. Wouldn't it make more sense to do the face capture in a separate pass? This may allow the performer to act more naturally and do multiple takes more easily. Trying to capture the entire performance (face, dialog, and body) in one perfect take may wind up being more difficult and complicated than you think, and the result could limit your options in post.
Regarding Kinect RGB for face capture: The Kinect Sensor has a fairly low framerate and resolution, which I'm sure will affect the quality of the face capture. I imagine the device will have to be somewhat close to your head to record suitable data for Zign Track or Maskarad, in which case it's probably way too close for full body capture, which may defeat the purpose for using Kinect.
This is all just my opinion of course. If anybody here can make this work, I'm very interesting in hearing about it, so please post. Thanks in advance for any info!
G.
