in collaboration with Metaio, Garching, Germany.
In this project, we are interested in the Implementation of an AR application for robust marker detection
- Important issues of marker detection system
Marker design - the application is currently working with bitonal square markers, which have two components:
- Low false positive detection rate
- Low inter-marker confusion
- Stable to different illumination changes and to partial occlusions
- Minimum marker size and maximum marker ID encoded<\li>
- Real-time performance
- Kernel - 6x6 bitonal pattern, which is sampled into 36-bit digital code
- Border – bi-tonal border, which is used for the proper 3D pose estimation of the marker.