Best solution for Kinect + IP Cameras + Resolume?

[kinwecamz]

Greetings from Europe! I’ve been lurking for a while and decided to register, because I would welcome some help with a project I’m working on. I’ve been searching the forums and looking into what can VVVV do and even though there are some answers for parts of what I want to do, and it seems VVVV should be capable of everything, I couldn’t find interestingly a topic that would cover all points.

What I would hope for is to get feedback on how to proceed, maybe I won’t even need Resolume, and ideally find guidance of someone who would help me with the patch. It would be nice to explore from beginning to end on my own, but frankly it is quite overwhelming even with all the documentation and I don’t really have the time to do so right now. Thus it would be nice to get some boost.

What I want to do?

I have direct links to live online IP cameras
These streams are usually mpjeg format
I have a Xbox 360 Kinect with Windows 7
I need these streams to switch based on position (how far/near the user is)

Originally I thought I would need Resolume and send the data and map the positions there, but since it seems VVVV can work with IP cameras and Kinect from the other threads I looked into directly, maybe I won’t.

How I imagine the solution?

I have the Kinect connected
I setup the axis
I extract the values for several positions
Each position would be for displaying a different camera
If position/value is detected play/switch camera

Before now I thought of having to use Resolume and XSplit, also Delicode and whatnot, but it seems VVVV is more effective. I’ve heard of Spout, but haven’t figured out how to use it.

The threads I looked at that might help were from karistouf, leonida, ft.

The possible solutions I looked into: FFGL, DXTory, Rootava, Mpejg texture, Osc protocol, DirectX FreeFrame, MidiOSC Mapper, Launchpad

Thanks!

[id144]

Hi @kinwecamz!

Try Kinect Hitboxes DX11 | vvvv , it’s a simple tool made for designers to create location specific interactions. Patch contains link to multiple sample projects.

Generally, when I start to work on a project, I break down the task into small segments - individual technical principles and I prototype them in a small separated patches. In you case, try to receive the online camera stream, try to receive two stream and switch between them, create sample that is switching between two to four values (user not present, user in zone 1, zone 2, user outside of interactive zone)

[kinwecamz]

Hi @id144, thanks for the link. I looked into that and while I was slightly overwhelmed by the options, I don’t think it was exactly what I had in mind - or at least I’m not sure how to transform it to my liking.

Your suggestion on how to proceed though is valid. I do have one patch for the cameras that I found in a different thread and one for the Kinect that I’ve made. Now I just need to make the switches in between and properly connect the two.