How Can I Interact?: Comparing Full Body Gesture Visualizations

This paper is dedicated to the question "How can I interact?", which may arise during a full body interaction game. To answer this question, a game needs to tell the players what actions are available and how those actions can be triggered. We focus on the video channel and use onscreen symbols to visualize how available input gestures have to be performed. We describe three symbol variants using recordings of a real person: color images, tracking shapes and skeletons, and solely tracking skeletons. An initial evaluation study shows clear advantages for the color images. We further outline how we extend the current implementation, for both improving the usability of the symbols, as well as easing their development.