Line-based Robust SfM with Little Image Overlap

Yohann Salaün (1, 2) , Renaud Marlet (2, 3) , Pascal Monasse (2)

1 Ecole Centrale Paris
2 LIGM - Laboratoire d'Informatique Gaspard-Monge
3 imagine [Marne-la-Vallée]

Usual Structure-from-Motion (SfM) techniques require at least trifocal overlaps to calibrate cameras and reconstruct a scene. We consider here scenarios of reduced image sets with little overlap, possibly as low as two images at most seeing the same part of the scene. We propose a new method, based on line coplanarity hypotheses, for estimating the relative scale of two independent bifocal calibrations sharing a camera, without the need of any trifocal information or Manhattan-world assumption. We use it to compute SfM in a chain of up-to-scale relative motions. For accuracy, we however also make use of trifocal information for line and/or point features, when present, relaxing usual trifocal constraints. For robustness to wrong assumptions and mismatches, we embed all constraints in a parameter-less RANSAC-like approach. Experiments show that we can calibrate datasets that previously could not, and that this wider applicability does not come at the cost of inaccuracy.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Yohann Salaün, Renaud Marlet, Pascal Monasse. Line-based Robust SfM with Little Image Overlap. 3DV 2017 International Conference on 3D Vision 2017, Oct 2017, Qingdao, China. ⟨10.1109/3DV.2017.00031⟩. ⟨hal-01618686⟩