Microsoft has just the right resource to at least point devs in the right direction, especially if they require a tad of help to get started.
“Building Great Gesture Detection” is a PowerPoint presentation available for download free of charge from Microsoft.
Authored by Claude Marais, Sr. Software Development Engineer Advanced Technology Group Microsoft, the overview stresses the most important aspects of marrying games with Kinect’s natural user interface.
“Gesture detection plays a fundamental role in Kinect titles since it is used to communicate player intent in a natural way. Determining if a player has performed a gesture seems trivial, but developing reliable gesture detection for even simple gestures, such as jumping or a hand wave, is quite challenging,” Microsoft stated.
This presentation covers case studies of building gesture detection, what has been learned so far, and recommendations for building your own gesture detection.
Devs will be able to get insight into three test cases from the PP presentation: Algorithmic: Wave Detection; Weighted Networks: Jump Detection and Exemplar Matching: Tennis Swing Detection.
Marais stresses that in order to build great gesture detection for Kinect, developers will need to perform consistent data collection, adapt their code to the data, and constant testing to ensure that the game’s response provides an excellent user experience.
What is Gesture Detection or Gesture Recognition? Computer’s ability to understand human gestures as input First used in 1963 with pen-based input device, Marais revealed.
In the end, gestures are nothing more than a form of communication, and in the specific case of Kinect for Xbox 360 a paradigm of human – computer interaction based on an input method that comes naturally to every user.