Right now I'm using 2 Kinect V2's with a trial license, I am noticing it is having a bit of trouble with fast moving animations like running, and with things like crouching.

I am hoping I made the right decision with going depth sensor route, but I just read that you can record up to a 20 foot space with the PS Eye, and after watching a video the guy looked like he was getting good tracking while running.

Have any of you tried both ways, and what were your results?

I could have got 16 PS eyes for the price of my 2 Kinects, I just want to know if I made a mistake.

...

They are 2 very different set ups, Kinects are much easier to set up and use, especially if area is a concern, but far less accurate, to me anyway, I absolutely can not use dual Kinects for the dance motions I record.

PS Eyes are a much more involved set up with much more critical specifications to follow for best results, but you can get a larger capture volume if you have the room.

There are plenty of references and opinions in the forum for both systems, you can use the search function to find many.

To each their own preference, but I personally chose 6 PS Eyes over dual Kinect v2, I have used both systems.

You may not like how critical it is to follow the set up and recording procedures for the PS Eyes, especially if you have to set up and break down after every session, but you will get better tracking, ability to have more realistic motion with less clean up and far less occlusion issues.

As far as 16 cameras, that's all good if you need that many, but better to only use the amount of cameras needed for a particular recording session, or if you are trying to capture 3 performers at once, more cameras will be mandatory.

PS Eye recordings are much more GPU intense and the tracking processing time will be much longer, all depending on the components in your processing machine.

Only way to really know for yourself is to test out 6 PS Eyes, compared to 2 Kinects and see if it fits in with your expectations, both systems have pros and cons.

...

Thanks for the reply, I don't think I could have got a better one.

It definitely helps put it into perspective, and I think the extra work setting up would definitely be worth it, especially since I sometimes have to record something 6 times which ends up being a load of extra work, before I'm happy, at least with fast movements.
But yes I guess I'm happy, right now I don't need a lot of space, and will mostly be using it for things like body gestures in cut scenes, which works pretty well.

So thanks again, really informative.

I'm just about to purchase a new system for my first mocap set up..

as far as my computer set up, I think it's a decent one (with i7 6850k and Fours gtx1070), that hopefully will benefit while recording, but I'm still not sure if I should go with dual Kinectv2 or 6 PS eye cams. From what snapz says, for faster motions one MUST use the PS cams.. so I want to be safe by going that route so that I don't regret it..

Can you post a sample bhv or fbx data that you made using your dual Kinectv2 set up? Also it would be super nice if someone with a 6 PS eye set up post a sample bhv or fbx data (dance perhaps) so that I can apply them in Maya (via hik) to a custm Daz model.. (I would render them with redshift3d -gpu renderer used for movies). ; )

I really need to decide quick which way I should go... : ) PS eyes or Kinectv2..



thanks