Globally Optimal 2D-3D Registration from Points or Lines without Correspondences
Authors:
- Mark BrownMBMark Brown
- David WindridgeDWDavid Windridge
- Jean-Yves GuillemautJGJean-Yves Guillemaut
- 1 December 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2111-2119
Abstract
We present a novel approach to 2D-3D registration from points or lines without correspondences. While there exist established solutions in the case where correspondences are known, there are many situations where it is not possible to reliably extract such correspondences across modalities, thus requiring the use of a correspondence-free registration algorithm. Existing correspondence-free methods rely on local search strategies and consequently have no guarantee of finding the optimal solution. In contrast, we present the first globally optimal approach to 2D-3D registration without correspondences, achieved by a Branch-and-Bound algorithm. Furthermore, a deterministic annealing procedure is proposed to speed up the nested branch-and-bound algorithm used. The theoretical and practical advantages this brings are demonstrated on a range of synthetic and real data where it is observed that the proposed approach is significantly more robust to high proportions of outliers compared to existing approaches.Keywords
This publication has 21 references indexed in Scilit:
- Line Matching and Pose Estimation for Unconstrained Model-to-Image Alignment2014
- Go-ICP: Solving 3D Registration Efficiently and Globally Optimally2013
- Incremental Line-based 3D Reconstruction using Geometric Constraints2013
- Matching 2D image lines to 3D models: Two improvements and a new algorithm2011
- Global Optimization through Rotation Space SearchInternational Journal of Computer Vision, 2009
- LSD: A Fast Line Segment Detector with a False Detection ControlIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Simultaneous pose and correspondence determination using line features2003
- SoftPOSIT: Simultaneous Pose and Correspondence Determination2002
- Solution of the Simultaneous Pose and Correspondence Problem Using Gaussian Error ModelComputer Vision and Image Understanding, 1999
- Random sample consensusCommunications of the ACM, 1981