Hierarchical ArUco Marker Array for Coarse-to-Fine Localization in XR applications

Summary

Some markers have been proposed to determine the positional relationship between an object and a camera for XR applications. Most of markers consider only a single scale, but there are many cases where the distance to the object varies greatly, and therefore, multi-scale recognition is required in XR systems. In this study, we propose a hierarchical marker array, which is composed of toned ArUco markers and forming fractal structure as shown in Fig. 1. The proposed marker array realizes handling multiscale and coarse-to-fine localization while using the conventional ArUco marker detection and identification algorithm without any modification. When the scale is changed in the actual scene or blurred by pre-processing, the array automatically shows the corresponding markers as shown in Fig. 2.

Fig. 1 Hierarchical ArUco Marker Array

Fig. 2 Selective detection of a certain layer by changing blur amount

References

1. Leo Miyashita, Satoshi Tabata and Masatoshi Ishikawa: Hierarchical ArUco Marker Array for Coarse-to-Fine Localization in XR applications, IEEE International Conference on Artificial Intelligence & eXtended and Virtual Reality (AIxVR2025), pp.209-212, Lisbon, Portugal, 27-29 Jan. (2025)