This work focuses on estimating an accurate 3D transformation in real time, which is used to register images acquired from different viewpoints. The main challenges are significant image appearance differences, which originate from lateral displacements and parallax, inconsistencies in our 3D model and achieving real-time execution. To this end, we propose a featurebased method using a single synthesized view, which can cope with significant image appearance differences. The 3D transformation is estimated using an EPnP refinement to minimize the influence of inconsistencies in the 3D model. We demonstrate that the proposed method achieves over 95% transformation accuracy for lateral displacements up to 350 cm, while still achieving 85% accuracy at displacements of 530 cm. Additionally, with a running time of 100 milliseconds, we achieve real-time execution as a result of efficiency optimizations and GPU implementations of time-critical components.
Dennis W.J.M.A van de Wouw, Martin A.R. Pieck, Gijs Dubbelman, Peter H.N. de With, "Real-time Estimation of the 3D Transformation between Images with Large Viewpoint Differences in Cluttered Environments" in Proc. IS&T Int’l. Symp. on Electronic Imaging: Image Processing: Algorithms and Systems XV, 2017, pp 109 - 116, https://doi.org/10.2352/ISSN.2470-1173.2017.13.IPAS-209