Corpus ID: 237532253

Rotation Averaging in a Split Second: A Primal-Dual Method and a Closed-Form for Cycle Graphs

@article{Moreira2021RotationAI,
  title={Rotation Averaging in a Split Second: A Primal-Dual Method and a Closed-Form for Cycle Graphs},
  author={Gabriel Moreira and Manuel Marques and Jo{\~a}o Paulo Costeira},
  journal={ArXiv},
  year={2021},
  volume={abs/2109.08046}
}
A cornerstone of geometric reconstruction, rotation averaging seeks the set of absolute rotations that optimally explains a set of measured relative orientations between them. In spite of being an integral part of bundle adjustment and structure-from-motion, averaging rotations is both a nonconvex and high-dimensional optimization problem. In this paper, we address it from a maximum likelihood estimation standpoint and make a twofold contribution. Firstly, we set forth a novel initialization… Expand

Figures and Tables from this paper

HARA: A Hierarchical Approach for Robust Rotation Averaging
We propose a novel hierarchical approach for multiple rotation averaging, dubbed HARA. Our method incrementally initializes the rotation graph based on a hierarchy of triplet support. The key idea isExpand

References

SHOWING 1-10 OF 26 REFERENCES
Rotation Averaging with the Chordal Distance: Global Minimizers and Strong Duality
TLDR
This paper analytically proves that in many cases there is no duality gap unless the noise levels are severe and proposes an efficient, scalable algorithm that outperforms general purpose numerical solvers by a large margin and compares favourably to current state-of-the-art. Expand
Rotation Averaging and Strong Duality
TLDR
This paper analytically prove that in many cases there is no duality gap unless the noise levels are severe, and proposes an efficient, scalable algorithm that outperforms general purpose numerical solvers and is able to handle the large problem instances commonly occurring in structure from motion settings. Expand
Shonan Rotation Averaging: Global Optimality by Surfing SO(p)n
TLDR
This work employs semidefinite relaxation in order to recover provably globally optimal solutions of the rotation averaging problem, and shows how to solve large-scale instances of these relaxations using manifold minimization on higher-dimensional rotation manifolds, re-using existing high-performance structure-from-motion pipelines. Expand
Robust Rotation and Translation Estimation in Multiview Reconstruction
  • Dan Martinec, T. Pajdla
  • Computer Science
  • 2007 IEEE Conference on Computer Vision and Pattern Recognition
  • 2007
TLDR
It is shown that iterative removal of pairwise reconstructions with the largest residual and reregistration removes most non-existent epipolar geometries. Expand
Spectral Synchronization of Multiple Views in SE(3)
TLDR
This paper addresses the problem of rigid-motion synchronization in the Special Euclidean Group SE(3), which finds application in structure-from-motion and registration of multiple three-dimensional point-sets and demonstrates through experimental results that the approach is significantly faster than the state of the art, while providing accurate estimates of rigid motions. Expand
A Riemannian low-rank method for optimization over semidefinite matrices with block-diagonal constraints
TLDR
A new algorithm is proposed to solve optimization problems of the form min f (X) for a smooth function f under the constraints that X is positive semidefinite and the diagonal blocks of X are small identity matrices and this method is compared against state of the art, mature software and finds that it is orders of magnitude faster, is more accurate and scales better. Expand
Novel Parameterization for Gauss–Newton Methods in 3-D Pose Graph Optimization
TLDR
This article derives an efficient implementation of the Gauss–Newton method for solving the rotation synchronization problem, and develops a convergence theory thereof, which shows that a structural parameter appears in the proof, which has a significant influence on the convergence basin. Expand
Fast Pose Graph Optimization via Krylov-Schur and Cholesky Factorization
TLDR
This work proposes a highly efficient algorithm to solve PGO which leverages the sparsity of the data by combining the Krylov- Schur method for spectral decomposition with Cholesky LDL factorization and allows for high scalability, low computational cost and high precision, simultaneously. Expand
GPSfM: Global Projective SFM Using Algebraic Constraints on Multi-View Fundamental Matrices
TLDR
This paper provides a complete algebraic characterization in the form of conditions that are both necessary and sufficient to enabling the recovery of camera matrices, and proposes a concrete algorithm for projective structure-from-motion that utilizes this characterization. Expand
Cramér–Rao Bounds and Optimal Design Metrics for Pose-Graph SLAM
TLDR
This article focuses on the relation between the graphical structure of pose-graph SLAM and Fisher information matrix, Cramér–Rao lower bounds (CRLB), and its optimal design metrics (T-optimality and D- Optimality). Expand
...
1
2
3
...