1.one kinect
2.2 sony ps eye cameras

can i use combinations of kinect and sony ps eye camera?

Hello,

No, you can't combine de 2.

If you have to do simple movement in front of the camera, you have not much place, kinect is the way.

If you want to do do more complicated movement, ps cameras are much much better.

The problem with kinect is that every parts of your body must be visible. You can't, for exemple do a profile walk, because de half part of your
body is not captured. This is very limitative. Another think, the maximum lenght from you to kinect is 5m... In other hands, if you have to do same simple movement, kinect capture very well. And with no calibration or setup.

With sony camera, you can do a lot more, with a simple 4 cameras setup. You must have a big empty space, calibration is much harder, but when you're ready... this is so cool... You can do every movement you want. Just awesome.

one kinect is better than two sony eye cams.

Just my opinion:

1 Kinect is better than 1 or 2 PS3 Eyes.

3 PS3 Eye cameras can be better than 1 Kinect depending on what you're trying to capture and track. The range of motions might be comparable but you get a larger capture space.

4 to 6 PS3 Eye cameras are significantly better than 1 Kinect. You can capture far more sophisticated motions this way plus you get a very large capture space.

That said, you must also consider the added complexity to using multiple PS3 Eye cameras vs. Kinect. Multiple PS3 Eye cameras require quite a bit more planning, several accessories, a more powerful computer, and a lot bigger space to record in. The extra work may be worth it as this allows you to to capture difficult motions that would be absolutely impossible with a single Kinect.

A final consideration: dual Kinect support may be coming soon to iPi DMC, which is supposed to be comparable to using 4 PS3 Eye cameras. It's too early to say how this really compares in practice since nobody outside of iPi Soft has tested it, but I suspect the capture space will still be smaller than four to six PS3 Eye cameras. Two Kinects may certainly broaden the kinds of motion you can capture though. All that said, two Kinects cost a bit more than six PS3 cameras. But then again, it will probably feature 'plug-and-play' simplicity like the current Kinect setup, which is very appealing.

Sigh! So many things to think about. :)

G.

_________________
Greenlaw
Artist/Partner - Little Green Dog | My Demo Reels (2013,) (2015,) (2017,) and (2019)

Watch a one minute excerpt on Vimeo now!

Time to update this info, now that we have support for 2 Kinects:

two Kinects are about as accurate as four PS3 Eye cameras.

Advantage: two Kinects are much easier to setup and calibrate than multiple PS3 Eye camera.

Disadvantages: 1.) two Kinects are more difficult to setup and calibrate than a single Kinect, 2.) has much smaller capture space than PS3 Eye cameras, 3.) motion blur with Kinect is more problematic than with PS3 Eye (Kinect is 30fps; PS3 Eye is 60dps.)

IMO, casual and beginner users will want a single Kinect setup; pro users and serious hobbyists will want 2 Kinects and/or 4 PS3 Eye cameras. Pro Users and serious hobbyists who don't have a lot of space and don't want to fuss with technical details will probably stay with two Kinects (most practical and flexible choice.) Pro users who intend to use iPi DMC on larger scale projects will probably want both systems, and if they have the space they will probably favor six PS3 Eye cameras.

So many options...but I'm not complaining. :)

G.

_________________
Greenlaw
Artist/Partner - Little Green Dog | My Demo Reels (2013,) (2015,) (2017,) and (2019)

Watch a one minute excerpt on Vimeo now!

thanks for info.

what about processing time? which one is faster?

This depends on your graphics card. I have a modest (by today's standard) Geforce GTX 460, and tracking PS3 Eye or Kinect data, single or dual, takes about 0.6 to 0.7 second per frame. Pretty fast IMO.

G.

_________________
Greenlaw
Artist/Partner - Little Green Dog | My Demo Reels (2013,) (2015,) (2017,) and (2019)

Watch a one minute excerpt on Vimeo now!

i have i3 and GForce GT 320 M. software uses 2 core. one of them as %70 second as %50. it is 3 secs per frame. i have the latest trial version. why it doesnt use 4 cores as %100 to calculate faster?

Because in your case bottleneck is GPU rather than CPU.

is there any way to use just cpu?