Comprehending and interpreting camera images (»Computer Vision«) is a core technology for virtual technologies because high-quality cameras are integrated in current computer and smart phone platforms and because the increasingly high performance facilitates elaborate image processing methods in real time.
In Augmented Reality, the real environment is captured by the camera and overlaid with digitally linked information. For this purpose, the pose (position and orientation) of the camera must be registered in real time and synchronized with the virtual camera. For industrial application scenarios, those model-based tracking procedures are particularly relevant, offering a high level of stability even with unstable lighting conditions.
The reference models used for computer vision-based tracking are directly derived from the objects’ CAD data.