deletions | additions       diff --git a/The_innovation_behind_Kinect_hinges__.tex b/The_innovation_behind_Kinect_hinges__.tex  index 6543aff..956ce5e 100644  --- a/The_innovation_behind_Kinect_hinges__.tex  +++ b/The_innovation_behind_Kinect_hinges__.tex 
...
The innovation behind Kinect hinges on advances in skeletal tracking. In skeletal tracking, a human body is represented by a number of joints representing body parts such as head, neck, shoulders, and arms (see Figure 4a). Each joint is represented by its 3D coordinates. The goal is to determine all the 3D parameters of these joints in real time to allow fluent interactivity and with limited computation resources allocated on the Xbox 360 so as not to impact gaming performance. Rather than trying to determine directly the body pose in this high-dimensional space, Jamie Shotton and his team met the challenge by proposing per-pixel, body-part recognition as an intermediate step (see Figure 4b). Due to their innovative work, Microsoft honored the Kinect Skeletal Tracking team members with the 2012 Outstanding Technical Achievement Award (www.microsoft.com/about/technicalrecognition/Kinect-Skeletal-Tracking.aspx).   Figure 5 illustrates the whole pipeline of Kinect skeletal tracking. The first step is to perform  per-pixel, body-part classification. The second step is to hypothesize the body joints by finding a global centroid of probability mass (local modes of density) through mean shift. The final stage is to map hypothesized joints to the skeletal joints and fit a skeleton by considering both temporal continuity and prior knowledge from skeletal train data.The skeletal data captured from the Kinect for Windows SDK for each pose consisted of the 3D positions of 20 joints. The positions of shoulder and elbow joints relative to the trunk were used to measure the angles of shoulder flexion and abduction(in degrees), while the positions of the elbow and hand relative to the trunk were used to measure the angle of external rotation. Using the skeletal data from the Kinect for Windows SDK, the Kinect was found to be highly reliability for the measurement of shoulder angle in most poses. Highest accuracy was achieved for the measurement of shoulder angle in the abduction to 90◦ pose. This is consistent with a previous finding that the Kinect was accurate for the measurement of ROM during shoulder abduction. \cite{huber2015validity}