Search the Community
Showing results for tags 'Interactivity Kinect Sensors'.
Found 1 result
I would like to put together a way for video characters, rear projected onto a screen via Watchout, to interact with real people. I found this helpful thread relating to sensors: http://forum.dataton.com/topic/994-wo-sensor/?hl=sensor And I found this thread dealing a bit with Kinect: http://forum.dataton.com/topic/436-watchout-kinect/ Kinect, and other motion controllers, like the ones for Wii and Playstation, according to the Wikipedia entry, seem to be more versatile and even include speech recognition, though, not being a gamer, I have no idea how well they do what they do. But what I was envisioning was having certain human movements or even human speech be interpreted by Kinect and then, using "glue code" that takes what Kinect has recognized, "adapts the data, and forwards relevant information to Watchout using its network protocol" (Mike's words) which would then trigger an auxiliary timeline in Watchout where a video character's response would be played back accordingly.. Once the "video character" responded, the video would continue with that video character just standing there waiting for a response until a human actually did respond, at which point another auxiliary time line would be triggered in the same way, replacing the running one, as a response to the human, etc, etc. So I'd film perhaps hundreds, or maybe eventually thousands, of possible responses that might possibly be triggered. So my questions are: 1. In general, is this reasonable, or have I, as usual, strayed far off base again? 2. Is there a limit to auxiliary time lines? 3. Has anyone tried this who might suggest any pointers? Perhaps a better way? 4. Does the use of auxiliary time lines seem like a reasonable way for a video character to respond? (Confession: I've never needed to use auxiliary timelines or even Midi so all I know about them is what I've read in the manual and on the forum) Input is most appreciated. Thanks! Lloyd