AR url__ar-Augmented Reality

1 Open Source http://artoolkit.org/

2 http://www.artishock.com/portfolio-augmented-reality.html

3

augmented reality (augmented Reality, AR) is a technique that calculates the position and angle of the camera image in real time and adds the corresponding image, the goal of which is to put the virtual world in the real world and interact with it on the screen. This technology was first introduced in 1990. With the improvement of the computing power of portable electronic products, the use of augmented reality is more and more wide.

The AR system has three outstanding characteristics: ① the real world and the virtual world information integration, ② has the real-time interaction, ③ is adds the localization virtual object in the three dimensional scale space. AR technology can be widely used in military, medical, architecture, education, engineering, film and television, entertainment and other fields.

Category: 1, based on the visual ar;2, GPS and directional sensor ar.

the types of AR based on Vision can be classified into two categories: based on the Marker Based and the non based tag (markerless). The key technology of AR is the three-dimensional tracking of realistic target (3D tracking), and it is sometimes said to be a six degree degree of freedom Tracking (6DOF) problem.

Based on the logo can be divided into: fixed template classes and custom template classes.

Fixed Template: This model has been more mature bar, the famous Artoolkit/plus,artag,artag in the anti-occlusion is superior to Artoolkit, later Artoolkit/plus adopted should be time Artag method.

Custom Template: The principle and fixed template ar is roughly the same, the difference on the template detection, mainly through the detection of feature points, and then according to the corresponding characteristics of different frames of the location change to obtain [R|t] matrix, Bazar is this type, the other like the following link, With very little code to achieve this type of AR demo,http://www.morethantechnical.com ... ar-in-opencv-wcode/, this code uses the optical flow method for feature point tracking, the following link code, With the surf feature, http://morethantechnical.googlecode.com/svn/trunk/opencv_ar/, no matter what, in addition to the characteristics of different, coordinate system transformation principle and the previous article similar, Some functions are provided in OPENCV to easily implement coordinate transformation, such as SOLVEPNP and cvfindhomography.

Based on no sign ar: Masterpiece is Ptam/m,mixare, will be the future development direction of AR

Tracking technology can be roughly divided into two categories, one is based on feature tracking (Feature Based tracking), such as tracking from the input images extracted from the feature point, feature Edge and so on to complete the target three-dimensional motion estimation and tracking. The other is based on template tracking (Template Based tracking) that is, through matching part of the input image Gray-level map implementation tracking.

Feature-based tracking Feature Based tracking is representative of the University of Oxford Engineering Robotics Institute Ptam, see Georg Klein left on the university's homepage http://www.robots.ox.ac.uk/~gk/PTAM/ Can download the learning source.

    template-based tracing Template Based Tracking representative work is the work of the Computer Vision Laboratory at the Federal Polytechnic University in Lausanne, see Vincent Lepetit Home Series Real-time tracking articles  http://cvlab.epfl.ch/~lepetit/  also worth downloading to learn Vincent collaborators Stefan Hintersser homepage  http ://campar.in.tum.de/main/stefanhinterstoisser  on the source, most of which are based on OPENCV.                AR technology can be widely used in military, medical, construction, education, engineering, Film and television, entertainment and other fields.