TY - GEN

T1 - CONSAC

T2 - Robust Multi-Model Fitting by Conditional Sample Consensus

AU - Kluger, Florian

AU - Brachmann, Eric

AU - Ackermann, Hanno

AU - Rother, Carsten

AU - Yang, Michael Ying

AU - Rosenhahn, Bodo

N1 - Funding Information: Acknowledgements This work was supported by the DFG grant COVMAP (RO 4804/2-1 and RO 2497/12-2) and has received funding from the European Research Council (ERC) under the European Union Horizon 2020 programme (grant No. 647769).

PY - 2020

Y1 - 2020

N2 - We present a robust estimator for fitting multiple parametric models of the same form to noisy measurements. Applications include finding multiple vanishing points in man-made scenes, fitting planes to architectural imagery, or estimating multiple rigid motions within the same sequence. In contrast to previous works, which resorted to hand-crafted search strategies for multiple model detection, we learn the search strategy from data. A neural network conditioned on previously detected models guides a RANSAC estimator to different subsets of all measurements, thereby finding model instances one after another. We train our method supervised as well as self-supervised. For supervised training of the search strategy, we contribute a new dataset for vanishing point estimation. Leveraging this dataset, the proposed algorithm is superior with respect to other robust estimators as well as to designated vanishing point estimation algorithms. For self-supervised learning of the search, we evaluate the proposed algorithm on multi-homography estimation and demonstrate an accuracy that is superior to state-of-the-art methods.

AB - We present a robust estimator for fitting multiple parametric models of the same form to noisy measurements. Applications include finding multiple vanishing points in man-made scenes, fitting planes to architectural imagery, or estimating multiple rigid motions within the same sequence. In contrast to previous works, which resorted to hand-crafted search strategies for multiple model detection, we learn the search strategy from data. A neural network conditioned on previously detected models guides a RANSAC estimator to different subsets of all measurements, thereby finding model instances one after another. We train our method supervised as well as self-supervised. For supervised training of the search strategy, we contribute a new dataset for vanishing point estimation. Leveraging this dataset, the proposed algorithm is superior with respect to other robust estimators as well as to designated vanishing point estimation algorithms. For self-supervised learning of the search, we evaluate the proposed algorithm on multi-homography estimation and demonstrate an accuracy that is superior to state-of-the-art methods.

KW - cs.CV

UR - http://www.scopus.com/inward/record.url?scp=85089157079&partnerID=8YFLogxK

U2 - 10.1109/CVPR42600.2020.00469

DO - 10.1109/CVPR42600.2020.00469

M3 - Conference contribution

SN - 978-1-7281-7169-2

T3 - Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)

SP - 4633

EP - 4642

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

PB - Institute of Electrical and Electronics Engineers Inc.

ER -