Abstract

This thesis proposes to replace the conventional marker of Augmented Reality (AR) with the QR Code pattern and we design and implement a prototype AR system named QRAR (Quick Response Code with Augmented Reality) from scratch. The system has several advantages, compared with traditional black-square-boarder tags. These advantages are: (1) better error correction to reinforce occlusion immunity, (2) the ability of embedding much more information into the code, (3) users do not have to predefine markers and train their AR programs to recognize their markers in advance, and (4) It is also possible to modify (part of) information of makers without recompiling AR programs. In addition, we introduce Kalman filter to solve the marker occlusion issue. The prototype of AR system uses the QR decoder which feeds back the information contained in the QR Code. The other major part of tasks in our prototype is to track the tilt angle and position of the marker with three bull-eye patterns which are the significant features of QR Code, related to the real world coordinate system. After that, we use part of the information contained in the QR Code to determine which virtual object we are about to draw and use the transformation matrix to make the virtual object in line with the tilt angle and position of the marker in the real world coordinate system. Our proposed system makes it possible to have applications fetch dynamic data from DB servers in the Intranet/Internet. Based on this, we implement a new application which could show a passport owner’s 3D image, assuming that a QR Code is attached with the passport, and our AR application can archive the image, stored in an internet database, identified by the QR Code.

Full Text
Published version (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