3D Gesture classification with linear acceleration and angular velocity sensing devices for video games

► Compared gesture recognition accuracy among five machine learning algorithms. ► Tested Decision Trees, Bayesian Networks, Linear Classifier, Support Vector Machines, and an AdaBoost-based Classifier. ► Conducted a user study with 27 participants, resulting in a database of 17,890 gesture samples. ► Investigated how recognition accuracy results translate to a video game environment. ► Suggest strategies for game developers who want to use 3D gestures in video games.