Freeframe plugin.. kinect on Windows??

After some time of trying (in vain) to have a decent working feed from Kinect into isadora under Win XP, I finally update myself to Win 7, and I ask myself why it took so long...

Now that Spout is available and being supported by Isadora via ffgl, I also installed "kinect for windows runtime v1.7", and I am asking myself what to do next to have kinect into isadora.

I wanted to check out the links previously posted in this thread, but they are not working.

how to proceed?

All the best.-

camilozk

now I made the kinect work following the next tutorial

http://www.youtube.com/watch?v=m5uTH3S9P9g&src_vid=TDuGfAV6hDc&feature=iv&annotation_id=annotation_393952

I still have no idea to send kinect images to isadora via spout. :(

Skulpture

I would love to help you with this but I have only recently just got a windows machine... Ive still to instal isadora on it. It's all very new to me. Sorry!

I am sure Ryan (DusX) will help you out as soon as he can.

gavspav

I couldn't find a working link to the depthcam freeframe plug in.

You might have to use some other software to get the kinect images into Isadora via Spout.

You could use Touch Designer/Processing/Max/OF etc.

Probably very very easy using TD.

I'm keen to do this with the KinectV2 but for some reason the Spout plugins aren't showing up for me :-(

DusX

I don't have a kinect so I haven't played with any of these yet, sorry.

Maximortal

Probably with vvvv you could do it. vvvv support kinect and spout too

keftaparty

Hi,

It works with vvvv sending to isadora with spout.

I have Kinect 2 working like this, even on an optimus laptop, thanks to spout & nvidia drivers updates.

Mehdi

crystalhorizon

@keftaparty sounds great! As I am not familiar with vvvv, is there anything special in the setup regarding spout? What kind of image do you get? How is the depth image different, than the kinect 1? Is there also a difference regarding the distance you can use the kinect 2?

camilozk

I was this week able to receive Kinect 1 data into processing, and I also received instructions from the people of Spout, on how to transfer this data to Isadora.

I will make it available if somebody needs it.

I would like to try it also in vvvv and check out what works better, @keftaparty: could you share the mechanism to get do this?

cheers!

crystalhorizon

That informations would be great

keftaparty

Hi,

@feinsinn , I can't answer about difference between kinect 1 and 2, I have only a kinect 2.

Kinect 2 is still a bit tricky to get into isadora, because for now spout only offers a DX9 sender for vvvv. And Kinect2 nodes are DX11 only.

So you have to first share your kinect2 texture from vvvv to vvvv ( best way I found to convert from DX11 to DX9 ), and then you can share it via spout to isadora...

Maybe it's better to wait a dx11 sender so it's straight forward. I am sure it will come soon.

Even if I don't have one to test, Kinect 1, the situation should be much easier.

Be sure to use the last beta of vvvv please.

Using the helppatch ( pressing F1 ) of the kinect node Kinect (Device Microsoft) EX9, you can see that a pin called kinect runtime can be connected to eather RGB, depth etc... nodes.

These nodes output a texture that you can send via spout ( using spout's helppatch as a help ).

I can make an example patch for Kinect 1, but as I have no device to try, I need a candidate to try it before posting it...

Mehdi

crystalhorizon

Hi Mehdi, Thank you for the information. I have a kinect 1, so I am interested in the new features of the new kinect 2. I use it mainly in a theatre setup. So what be nice to know how the distance is different: the kinect one is only usable in a quite short distance for a theatre. Has the kinect 2 a wider range? Is there a link where one can be updated with the new developments of the for third party apps? Also if the different data can be used in Isadora in a near future. All the best, Alex

gavspav

I've got the kinect v2 into isadora via max/spout. The immediately noticeable differences are that the colour image is a much higher resolution and generally the camera has lower latency. There are more joints tracked and there is a feature which spots if a persons hands are opened or closed. There is also face tracking and more features but they are not all implemented yet.

camilozk

just a comparison table:

http://channel9.msdn.com/coding4fun/kinect/Kinect-1-vs-Kinect-2-a-side-by-side-reference