The C++ Augmented Reality Toolkit
Contents
1. Introduction 2. Previous and Current Work 3. Image Segmentation 4. Feature Extraction 5. Marker Detection 6. Camera Orientation 7. Using ARLib 8. Conclusion A. Sample Otsu Thresholding C++ Code B. ARLib Application Command-Line Switches C. ARLib Application User Interface Keys D. ARLib Configuration File References ARLib API Documentation2. Previous and Current Work
The Computer Vision Homepage [HUBER 2005] provides a list of known computer vision research institutes from around the world. At over 300 in September 2005, the number of entries in this list reveals just how much current interest and activity there is in the subject.
A key aspect of Computer Vision is the ability of the application to identify known entities within an image. Some techniques for performing computer vision use markers to aid in the recognition process whereas more recently markerless detection systems have been developed [VACCHETTI et al 2003], [ALLEZARD et al 2000] and [LEPETIT 2004]. These markerless systems still require some knowledge about the environment. Some techniques use a computer model of an object to use as a template, where lines and features detected in the scene are matched against the model’s shape. Other techniques search for textures on objects.
The application of Augmented Reality within the realms of Computer Vision is becoming more common and is now used in many different fields, including medical, manufacturing, military and visualisation. In particular, fighter pilots have been using the application of augmented reality for many years to highlight the outside world in the form of head up displays.
The University of North Carolina has researched into using augmented reality in medicine [LIVINGSTON 1998] as an aid for performing surgery, see figure 2.1.
![Figure 2.1<br/> Augmenting the internal organs during surgery [LIVINGSTON 1998].](images/arlib/figure-2-1.png)
Another interesting example of Augmented Reality is an application that was developed to aid maintenance engineers who have to repair a variety of different equipment [FEINER 1993]. The system works by incorporating a camera into the engineer’s goggles, which is linked to the AR software. The AR application is able to annotate what the engineer is seeing with useful text and markers, to aid the engineer in his job, see figure 2.2.
 |
 |
|
Figure 2.2 |
|
A mundane but essential task of quality control has been made easier by AR. Custard creams are scanned for excess cream by an AR application and rejected if they fall outside of a predefined limit [DAVIS 2005].
L'Ecole Polytechnique Fédérale de Lausanne (EPFL) is a Swiss University that has a dedicated computer vision lab. Within this lab they have developed many techniques and applications in the realms of augmented reality, including real-time face tracking and 3D object tracking.
 |
 |
|
Figure 2.3 |
|
It is EPFL’s 3D object tracking work that inspired the development of the ARLib, and is described in detail in the following chapters.
| << 1. Introduction | 3. Image Segmentation >> |