Snapz wrote:
...I will never use 4, actually I think 2 will perform very well, all I need it for would be a bit better rotation tracking...
Yes, two v2 devices should be significantly better than one v2 device.
In my experience with multi-Kinect v1, using two v1 devices is a HUGE improvement over using a single v1 device. As mentioned elsewhere, I used the dual system to create short visual effects shots for a handful of b-movies for SyFy channel during the 2013 - 2014 seasons. I'll see if I can post some of that footage with the setups for comparison.
I used the triple configuration on a couple shots and it improves the occlusion issue even further. But I think two devices is adequate for many positions. In fact, when I'm pressed for time, I still use two for many captures because it's faster to set up and faster to track. For many common poses/motions, you might not notice a big difference. I mainly use the triple Kinect system for difficult motions.
This year I'm back to using mocap only for our own short film productions, and what I'm hoping to gain from the quad Kinect configuration is smoother raw data and fewer tracking errors. That should speed up the process but I think the tradeoff could be that my old computer will take a bit longer to track the v2 data. We'll see.
If it turns out to be much slower, I'll just have to deal with it because I basically used my 'new desktop computer' budget for the upgraded mocap gear. :p
Stay tuned.
G.
