Joint Filtering: Enhancing gesture and mouse movement in Microsoft Kinect application

One of the most imperative research areas in the field of Human Computer Interaction (HCI) is gesture recognition as it provides a natural and spontaneous way to communicate between people and machines. Gesture based applications range from computer games to virtual augmented reality and is recently being explored in other fields. The idea behind this work is to attempt to tackle the problem of gesture movement of the human body using Microsoft Kinect sensor. The skeletal tracking (ST) system of the Natural User Interface (NUI) provides joint positions of tracked persons skeletons. These joint positions are the data consumed as position and pose, and they are used for many purposes, such as gesture detection, navigating user interfaces, and so on. The filter method used is Holt exponential smoothing filter. This filter smooth out the jitter, adds the capability to continue tracking for a short period of time when the subject moves out of range of the sensor, and improves the accuracy of the tracking by incorporating the information.