The Kinect Sensor hardware specification has been
described in many recent publications [3], [5]–[7]. The Kinect
Sensor (Figure 1) has a 640 x 480 32-bit RGB color camera
and a 320*240 16-bit depth camera both running at 30 frames
per second. The field of view (FOV) of cameras are 57º
horizontally and 43º vertically. It has a motorized angle driver
which can tile the both camera 27º vertically and a microphone
array which is used for detecting voice commands [8]. The
Kinect Sensor hardware specification is able to apply in many
applications such as medical, robotic, and gaming application.
The working principle of Microsoft Kinect Sensor is that
emission of IR pattern from Kinect depth sensor and
simultaneously image capture of IR images with a (traditional)
CMOS camera that is fitted with an IR-pass filter. The image
processor of Kinect Sensor uses the relative positions of the
dots in the pattern to calculate the depth displacement at each
pixel position in the image. The algorithm for calculating the
depth and accuracy of the Kinect Sensor has yet not released
by Microsoft. However, few papers have computed the
absolute accuracy of the depth map [9]–[11].The depth sensor
from Kinect Sensor returns an 11-bit number raw data which
needs further processing to get the true depth. The true depth is
expressed as found in [9]Khoshelham et al