Unsupervised Learning for Robust Fitting: A Reinforcement Learning Approach
Paper i proceeding, 2021

Robust model fitting is a core algorithm in a large number of computer vision applications. Solving this problem efficiently for datasets highly contaminated with outliers is, however, still challenging due to the underlying computational complexity. Recent literature has focused on learning-based algorithms. However, most approaches are supervised (which require a large amount of labelled training data). In this paper, we introduce a novel unsupervised learning framework that learns to directly solve robust model fitting. Unlike other methods, our work is agnostic to the underlying input features, and can be easily generalized to a wide variety of LP-type problems with quasi-convex residuals. We empirically show that our method outperforms existing unsupervised learning approaches, and achieves competitive results compared to traditional methods on several important computer vision problems.

Författare

Giang Truong

Edith Cowan University

Huu Le

Datorseende och medicinsk bildanalys

David Suter

Edith Cowan University

Erchuan Zhang

Edith Cowan University

Syed Zulqarnain Gilani

Edith Cowan University

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

10636919 (ISSN)

10343-10352
9781665445092 (ISBN)

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
Virtual, Online, USA,

Ämneskategorier

Beräkningsmatematik

Bioinformatik (beräkningsbiologi)

Datorseende och robotik (autonoma system)

DOI

10.1109/CVPR46437.2021.01021

Mer information

Senast uppdaterat

2022-02-01