Deep fundamental matrix estimation without correspondences
Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
Estimating fundamental matrices is a classic problem in computer vision. Traditional methods rely heavily on the correctness of estimated key-point correspondences, which can be noisy and unreliable. As a result, it is difficult for these methods to handle image pairs with large occlusion or significantly different camera poses. In this paper, we propose novel neural network architectures to estimate fundamental matrices in an end-to-end manner without relying on point correspondences. New modules and layers are introduced in order to preserve mathematical properties of the fundamental matrix as a homogeneous rank-2 matrix with seven degrees of freedom. We analyze performance of the proposed models using various metrics on the KITTI dataset, and show that they achieve competitive performance with traditional methods without the need for extracting correspondences.
Originalsprog | Engelsk |
---|---|
Tidsskrift | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Sider (fra-til) | 485-497 |
Antal sider | 13 |
ISSN | 0302-9743 |
DOI | |
Status | Udgivet - 2019 |
Eksternt udgivet | Ja |
Begivenhed | 15th European Conference on Computer Vision, ECCV 2018 - Munich, Tyskland Varighed: 8 sep. 2018 → 14 sep. 2018 |
Konference
Konference | 15th European Conference on Computer Vision, ECCV 2018 |
---|---|
Land | Tyskland |
By | Munich |
Periode | 08/09/2018 → 14/09/2018 |
Bibliografisk note
Publisher Copyright:
© Springer Nature Switzerland AG 2019.
ID: 301824797