A Graduated Filter Method for Large Scale Robust Estimation
Paper in proceeding, 2020

Due to the highly non-convex nature of large-scale robust parameter estimation, avoiding poor local minima is challenging in real-world applications where input data is contaminated by a large or unknown fraction of outliers. In this paper, we introduce a novel solver for robust estimation that possesses a strong ability to escape poor local minima. Our algorithm is built upon the class of traditional graduated optimization techniques, which are considered state-of-the-art local methods to solve problems having many poor minima. The novelty of our work lies in the introduction of an adaptive kernel (or residual) scaling scheme, which allows us to achieve faster convergence rates. Like other existing methods that aim to return good local minima for robust estimation tasks, our method relaxes the original robust problem but adapts a filter framework from non-linear constrained optimization to automatically choose the level of relaxation. Experimental results on real large-scale datasets such as bundle adjustment instances demonstrate that our proposed method achieves competitive results.

large-scale optimization

3D reconstruction

robust estimation

bundle adjustment

Author

Huu Le

Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering

Christopher Zach

Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering

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

10636919 (ISSN)

5558-5567 9157809

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition
Seattle, USA,

Subject Categories

Control Engineering

Signal Processing

Computer Vision and Robotics (Autonomous Systems)

DOI

10.1109/CVPR42600.2020.00560

More information

Latest update

11/25/2020