6D Camera Relocalization in Ambiguous Scenes via Continuous Multimodal Inference
Bui M, Birdal T, Deng H, Albarqouni S, Guibas L, Ilic S, Navab N (2020)
Publication Type: Conference contribution
Publication year: 2020
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 12363 LNCS
Pages Range: 139-157
Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Event location: Glasgow, GBR
ISBN: 9783030585228
DOI: 10.1007/978-3-030-58523-5_9
Abstract
We present a multimodal camera relocalization framework that captures ambiguities and uncertainties with continuous mixture models defined on the manifold of camera poses. In highly ambiguous environments, which can easily arise due to symmetries and repetitive structures in the scene, computing one plausible solution (what most state-of-the-art methods currently regress) may not be sufficient. Instead we predict multiple camera pose hypotheses as well as the respective uncertainty for each prediction. Towards this aim, we use Bingham distributions, to model the orientation of the camera pose, and a multivariate Gaussian to model the position, with an end-to-end deep neural network. By incorporating a Winner-Takes-All training scheme, we finally obtain a mixture model that is well suited for explaining ambiguities in the scene, yet does not suffer from mode collapse, a common problem with mixture density networks. We introduce a new dataset specifically designed to foster camera localization research in ambiguous environments and exhaustively evaluate our method on synthetic as well as real data on both ambiguous scenes and on non-ambiguous benchmark datasets. We plan to release our code and dataset under multimodal3dvision.github.io.
Involved external institutions
Germany (DE)
United States (USA) (US)How to cite
APA:
Bui, M., Birdal, T., Deng, H., Albarqouni, S., Guibas, L., Ilic, S., & Navab, N. (2020). 6D Camera Relocalization in Ambiguous Scenes via Continuous Multimodal Inference. In Andrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 139-157). Glasgow, GBR: Springer Science and Business Media Deutschland GmbH.
MLA:
Bui, Mai, et al. "6D Camera Relocalization in Ambiguous Scenes via Continuous Multimodal Inference." Proceedings of the 16th European Conference on Computer Vision, ECCV 2020, Glasgow, GBR Ed. Andrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm, Springer Science and Business Media Deutschland GmbH, 2020. 139-157.
BibTeX: Download