The typical registration process in augmented reality CAR) consists of three independent consecutive stages: static calibration, dynamic tracking, and graphics overlay. The result is that the real-virtual registration is “open loop”-inaccurate calibration or tracking leads to misregistration that is seen by the users but not the system. To cope with this, we propose a general approach to “close the loop” in the displayed appearance by using the visual feedback of registration for pose tracking to achieve accurate registration. Specifically, a model-based method is introduced to simultaneously track and augment real objects in a closed-loop fashion, where the model is comprised of the combination of the real object to be tracked and the virtual object to be rendered. This method is applicable to paradigms including video-based AR, projector-based AR, and diminished reality. Both qualitative and quantitative experiments are presented to demonstrate the feasibility and effectiveness of our approach.
No takes yet. Share an insight, caveat, or question.
Zheng et al. (2013) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: