Abstract
We present a family of methods for 2D–3D registration spanning both deterministic and non-deterministic branch-and-bound approaches. Critically, the methods exhibit invariance to the underlying scene primitives, enabling e.g. points and lines to be treated on an equivalent basis, potentially enabling a broader range of problems to be tackled while maximising available scene information, all scene primitives being simultaneously considered. Being a branch-and-bound based approach, the method furthermore enjoys intrinsic guarantees of global optimality; while branch-and-bound approaches have been employed in a number of computer vision contexts, the proposed method represents the first time that this strategy has been applied to the 2D–3D correspondence-free registration problem from points and lines. Within the proposed procedure, deterministic and probabilistic procedures serve to speed up the nested branch-and-bound search while maintaining optimality. Experimental evaluation with synthetic and real data indicates that the proposed approach significantly increases both accuracy and robustness compared to the state of the art.
Highlights
This paper deals with the general problem of 2D–3D registration where given an image taken by a calibrated camera and a 3D model, the objective is to determine the pose of the camera with respect to the model
In this paper we present a family of globally optimal solutions to the 2D–3D registration problem from points and lines without correspondences and in the presence of outliers
It is observed that the proposed globally optimal approaches perform significantly better than SoftPosit and RANSAC
Summary
This paper deals with the general problem of 2D–3D registration where given an image taken by a calibrated camera and a 3D model, the objective is to determine the pose of the camera with respect to the model. While there exist established solutions to this problem in the case where correspondences are known, there are many situations where it is not possible to reliably extract such correspondences across modalities, requiring the use of a correspondence-free registration algorithm. While there exist techniques to extract features in the 2D and 3D domains (e.g. corners [8], salient features [9] or lines [10,11]), it is an open problem to automatically establish correspondences between them. This may be explained by a variety of reasons. More generally, correspondences of any feature type are difficult to hypothesise when the 3D model is untextured, as is often the case if it is obtained by a laser range scanner
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
