DocumentCode
3746455
Title
A new interaction method for augmented reality based on ARToolKit
Author
Huibai Wang;Junli Qin;Fengquan Zhang
Author_Institution
College of Computer Science and Technology, North China University of Technology, Beijing, China
fYear
2015
Firstpage
578
Lastpage
583
Abstract
In this paper, we implement a new interaction method for the application of augmented reality using ARToolKit and OpenCV SDK. First, on the basis of the exiting marker-based tracking system, this paper achieved the multi-marker tracking that solved the tracking failure problem when the single marker is partially occluded, which greatly improves the robustness of the registration algorithm. In addition, Codebook-based foreground detection model is used to detect the hand region under an unexpected background environment. Moreover, we could recognize the hand gesture by convex hull on the basis of the hand contour. Therefore, a gesture recognition algorithm is presented based on codebook model algorithm and convex hull algorithm for the interaction. In the end, we use several different experiments to show the performance, which can be seen the good interactivity and realistic.
Keywords
"Image color analysis","Brightness","Solid modeling","Algorithm design and analysis","Cameras","Training","Three-dimensional displays"
Publisher
ieee
Conference_Titel
Image and Signal Processing (CISP), 2015 8th International Congress on
Type
conf
DOI
10.1109/CISP.2015.7407945
Filename
7407945
Link To Document