Difference between revisions of "Positional tracking"

From Virtual Reality, Augmented Reality Wiki
Jump to: navigation, search
Line 38: Line 38:
 
===Sensor Fusion===
 
===Sensor Fusion===
 
Sensor fusion is when multiple tracking methods are used. The different methods can cover each other's weaknesses.
 
Sensor fusion is when multiple tracking methods are used. The different methods can cover each other's weaknesses.
 +
 +
 +
[[Category:Terms]]

Revision as of 09:43, 12 April 2015

Positional tracking tracks the absolute position of the HMD, hands and other devices and body parts within the 3 dimensional space. Positional tracking records the translational movements, left/right, forward/backward and up/down, of your device. Precise and low latency positional tracking is essential to have a good VR experience and produce presence.

By contrast, rotational tracking, records the rotational movements of your HMD, such as turning and tiling of your head.

Methods

Acoustic

Acoustic tracking measures the length of time of an acoustic signal takes to reach various receivers.

Acoustic tracking can be influenced by surrounding noise causing inaccurate measurements. It also demands careful calibration and does not provide high update rates. Even with its faults, acoustic tracking can be used injunction with other methods to improve positional tracking.

Inertial

Inertial tracking is performed by accelerometer and gyroscope. Accelerometer measures the linear acceleration while gyroscope measures the angular velocity.

While accelerometer and gyroscope supply low latency and high update rate. They do not provide positional information accurately.

Magnetic

Magnetic tracking determines the strength of the magnetic field in different directions.

Magnetic tracking is utilized in Razer Hydra. It is usually a very accurate tracking method, although it can be influenced by magnetic fields generated by other electronics and objects near the sensor or emitter.

Most HMDs and smartphones contain magnetometers, IMUs that detect the magnetic field of Earth.

Optic

Marker

Optical tracking with mark consist of adding an unique marker onto an object you want to track. Then use a camera or multiple cameras to follow the movement and position of the marker. Finally algorithms are used to figure out the position and orientation of the object.

The marker can be active or pass. Active marker, such as an IR light, can flash periodically to sync with the camera. A retroreflector can be used as a passive marker to reflect IR light emitted by the camera to create the same effect as a active one.

Multiple markers with unique designs can be deployed. They allow the camera(s) to track the positions of multiple objects at the same time.

Markerless

Certain identifiable features, such as the fingers on your hands, can be tracked by camera(s) even without marker(s). Leap Motion employs this to find the position and movement of your fingers.

Depth Map

Depth Map camera constructs a live view of the distances of the objects from the camera.

Microsoft's Kinect uses depth map camera.

Sensor Fusion

Sensor fusion is when multiple tracking methods are used. The different methods can cover each other's weaknesses.