Here´s some nice info on how Ubisoft does it: http://www.youtube.com/watch?v=jzkSNCeHuSA
You may also find Marc Schoenagel´s blog (Softimage Evangelist) has a video of him playing with a Microsoft kinect and real-time capture directly into the Softimage viewport. There´s also someone on the list here who has written a compareable plugin that even has the advantage of not having to have the timeline playback active. I´ve forgot his name, unfortunately. Toying with the XBox Kinect is on my personal to-do list. Especially for 3D scans. It doesn´t matter if you can clean and refine scans coming from LIDAR or a Kinect... As I said, you probably looked into Motionbuilder already. Cheers, tim On 16.08.2012 08:10, Mirko Jankovic wrote:
On topic of learning in project. One thing is to learn an part of thing that you are missing in your tool set, but it is completely different thing to learn whole process on project in software that you never touched. That will slow down production almost to a halt and can create a lot of problems once you do things in a wrong way so to speak in one part of production that comes back to haunt you later on. So again, one thing is learning for example how to implement Facerobot in your pipeline when you already set pipeline in your software, but it is completely different when you meat with Si interface and workflow for the first time and wanna learn all at once. That is why I mentioned that it is better to break up production and tackle one thing at the time.
