Forums > MaxMSP

problems with jit.openni tracking individual points

April 2, 2013 | 2:06 pm

Hi everyone, just a quick question which I’m sure there is an obvious solution to, I just can’t work it out.

I am using [jit.open.ni] for mac, which is pretty amazing. I’m working with groups of autistic kids and need to be able to track them individually, automatically isolating the individual closest. This I can do – in theory – but I’m getting stuck on something.

Lets look at the mapping of the right hand as an example [OSC-route /*/r_hand /1/r_hand /2/r_hand /3/r_hand /4/r_hand]

Is [*/r_hand] an aggregate of the other users or a user in its own right? It’s weird because sometimes it acts as an individual user i.e it spits out data when none of the other routings ([/1/r_hand /2/r_hand] etc.) are not. But at other times, when more than one user is present, the data it spits out erratically jumps between values as if it was aggregating data. Ideally I would like to totally ignore the [/*/r_hand] and concentrate on the other outputs, but as I’ve mentioned this is not possible because sometimes the [/*/r_hand] acts as a user.

Any help would be appreciated!

p.s what is Scene Floor Data? Is it the floor?!


July 1, 2013 | 3:32 pm

I only today saw this. You are asking an OSC question. I recommend you study the help file that comes with OSC-route and learn from it + google. Then reference https://github.com/diablodale/jit.openni/wiki#skeleton-joint-data

openni identifies users aka humans by numbers. When you use the * (the wildcard) you can getting any and all users. OSC-route is like a search engine. by using /*/r_hand, you can asking OSC-route to find any OSC message that starts with anything (aka /*) followed by "/r_hand"


Viewing 2 posts - 1 through 2 (of 2 total)