Answers are grounded in this paper and cite their sources.
This evaluation compares tracking accuracy in virtual and physical environments, suggesting enhancements for AR/VR systems.
Robust object tracking is a foundational technology for augmented and virtual reality (AR/VR) systems. While many benchmarking efforts rely solely on simulated environments, these often lack real-world fidelity, limiting practical insights. To bridge this gap, we propose a hybrid evaluation approach, combining a physical test rig-constructed from aluminum extrusion frames and ArUco markers with a digital twin created in Blender. Our approach facilitates direct sim-to-real comparisons of fiducial marker tracking performance. We benchmark the widely used OpenCV ArUco 'Augmented Reality library from the University of Cordoba' against a custom-optimized tracking approach, BART (Bielefeld Augmented Reality Tracker), a modern performance-optimized alternative, to evaluate whether tracking limitations stem from the ArUco implementation itself or are inherent to fiducial marker tracking in general. Our evaluation focuses on detection accuracy and pose estimation performance. The results offer practical insight into the differences between tracking performance in real and simulated environments.
No takes yet. Share an insight, caveat, or question.
Krumstroh et al. (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: