Robust synchronization in SO(3) and SE(3) via low-rank and sparse matrix decomposition

• Published in: Computer vision and image understanding Vol. 174; pp. 95 - 113
• Main Authors: Arrigoni, Federica, Rossi, Beatrice, Fragneto, Pasqualina, Fusiello, Andrea
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.09.2018
• Subjects: Absolute rotations; Global registration; Global rotations; Low-rank & sparse matrix decomposition; Matrix completion; Robust principal component analysis; Structure-from-motion; ℓ1-Regularization; Absolute rotations; Global rotations; Robust principal component analysis; Global registration; Low-rank & sparse matrix decomposition; ℓ1-Regularization; Structure-from-motion; Matrix completion
• ISSN: 1077-3142; 1090-235X
• DOI: 10.1016/j.cviu.2018.08.001

•Synchronization in SO(3) and SE(3) is formulated as a low-rank and sparse matrix decomposition problem.•Any low-rank and sparse matrix decomposition algorithm can be used in this framework.•Good trade-off between resistance to outliers and speed. This paper deals with the synchronization problem, which arises in multiple 3D point-set registration and in structure-from-motion. The problem is formulated as a low-rank and sparse matrix decomposition that caters for missing data, outliers and noise, and it benefits from a wealth of available decomposition algorithms that can be plugged-in. A minimization strategy, dubbed R-GoDec, is also proposed. Experimental results on simulated and real data show that this approach offers a good trade-off between resistance to outliers and speed.