Abstract
Augmented reality (AR) is a field of computer research which deals with the combination of real world and computer generated data. Registration is one of the most difficult problems currently limiting the usability of AR systems. In this paper, we propose a novel natural feature tracking based registration method for AR applications. The proposed method has following advantages: (1) it is simple and efficient, as no man-made markers are needed for both indoor and outdoor AR applications; moreover, it can work with arbitrary geometric shapes including planar, near planar and non planar structures which really enhance the usability of AR systems. (2) Thanks to the reduced SIFT based augmented optical flow tracker, the virtual scene can still be augmented on the specified areas even under the circumstances of occlusion and large changes in viewpoint during the entire process. (3) It is easy to use, because the adaptive classification tree based matching strategy can give us fast and accurate initialization, even when the initial camera is different from the reference image to a large degree. Experimental evaluations validate the performance of the proposed method for online pose tracking and augmentation.
Highlights
The main intention of augmented reality is to superimpose extra perceptible elements on a user’s real world environment, for the purpose of improved understanding and interaction
We propose a novel feature tracking strategy combing wide and narrow baseline matching techniques for Augmented reality (AR) systems
Experimental results prove that the proposed method applies to both indoor and outdoor AR systems and is precise enough even in the cases of partial occlusion, rapid camera movement and large changes in volumes and viewpoints
Summary
The main intention of augmented reality is to superimpose extra perceptible elements on a user’s real world environment, for the purpose of improved understanding and interaction. Since vision plays an important role in human perception, most AR research is concerned with the use of live video imagery which is digitally processed and "augmented" by the addition of computer generated graphics. AR requires accurate registration of virtual objects in 3D in order to render a virtual object into the real world
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
