The course will first introduce the most effective practices of real-time body interaction based on computer vision, like movement detection and position tracking. This interaction will be used to influence or generate realtime audio-visual content. Later we will then explore more articulate modalities of interactions based on skeleton tracking, pose detection and hand gesture recognition. Hardware-wise we will use standard cameras/webcams/phone-cams, kinect sensors and leap motion sensors, software-wise we’ll use MAX (https://cycling74.com) and Wekinator (http://www.wekinator.org) an accessible Machine Learning software to help us map complex data to desired output parameters. As output we will generate and control sounds in real-time within MAX and communicate with Unity (https://unity.com) for the visuals.

If you own and are interested in using kinect sensors and/or leapmotion and/or cameras/webcams please bring them