I need ideas for my Kinect + Max/MSP project.

kinectfan's icon

Doing a gesture controlled interface in Max/MSP using the Kinect. I need somebody to help me finding an interesting approach to using Kinect + music. This is a bachelor thesis.

Here are some of my ideas:

- make a prototype to show a way of making live performance more interesting by gesture recognition to trigger effects on sounds/music

- make a prototype to show a way of using your body movements/hands to control effects in a sequencer instead of using your mouse or knobs on a synth/mixer/midi module

- more ideas? (KEYWORDS: kinect + sound + max/msp)

dtr's icon

bit weird that you start with the tools and then try to come up with the ideas, no?

anyway, here's my performance project in development, using Kinect motion tracking: http://dietervandoren.net/index.php?/project/integration03/

might inspire you one way or another

brendan mccloskey's icon

Hi
dtr's first comment is fundamental to an interaction design project at this level. Putting the Kinect to one side for a moment will allow you to assess exactly what the demands of the environment are. Ask yourself the following questions first:

-what are the expected abilities and input modalities required by the environment? Coarse or fine gestures, or both?
-what are the anticipated sound (media) events?
-how will the gestures be mapped to those events - directly or indirectly? One-to-one or one-to-many?

No doubt your supervisor will have pointed you towards Wanderley's Trends in Gestural Control of Music (IRCAM 2000) and Hunt/Kirk's papers on The Importance of Parameter Mapping (NIME 2002).

I think your first idea of gesture recognition would be the most fruitful - IMHO. Coupled to a live audio input stream this could be quite radical.

Best regards
Brendan

Thinksamuel's icon

Your way of thinking is totally wrong: You start with tools and develop a way to use them. that is not the right thing to do: YOu should first establish your goals on an artistic level and when you have formed your ideas firmly, you can start looking for tools/ toys. If you do not establish your goal first you will end up losing a lot of time.

brendan mccloskey's icon

@dtr

ps, very cool application of motion/gesture tracking!

Brendan

dtr's icon

@n00b_meister: tanx! the video is of the first experimental version. i'm working hard on making it something that can truly be called an instrument...

prof_lofi's icon

I don't agree, I think working with a new piece of kit/tech can inspire or push you to new artistic places. What I do find weird though is asking the group for 'ideas'. Maybe a better way to phrase it would be to ask how others have used it so far and use that as a catalyst. Maybe an aven better way is to commit to using it in a way that no one (to your knowledge) has used it for so far...then ask the group for help for code/execution etc.

Bill

balam's icon

also consider the limitation of the tools and environmental conditions.

ćwiek's icon
MIB's icon

@dtr: very cool!! loved the use of movement and light in both integration 3 + 4!! Do you have some documentation/blogs where one could get some further insight into your process?

dtr's icon

Not at the moment, sorry. There's a snippet in the video on http://www.dietervandoren.net/index.php?/project/integration04/ that gives a tiny glimpse of the Max stuff going on, not more. I might get to work on more documentation in the coming months. A nerdy tech report could be part of it ;)

trumpman00's icon

hey DTR is there anyway i can get this patch?

dtr's icon

'this patch' is a modular system of a dozen patches, some python code and a bunch of specific hardware (kinects, sensor gloves, beamers, audio/VGA/DMX interfaces etc). Not likely useful to anyone in its current form. Any element in particular you're interested in?