Deep Bingham Networks: Dealing with Uncertainty and Ambiguity in Pose Estimation
Deng H, Bui M, Navab N, Guibas L, Ilic S, Birdal T (2022)
Publication Type: Journal article
Publication year: 2022
Journal
Book Volume: 130
Pages Range: 1627-1654
Journal Issue: 7
DOI: 10.1007/s11263-022-01612-w
Abstract
In this work, we introduce Deep Bingham Networks (DBN), a generic framework that can naturally handle pose-related uncertainties and ambiguities arising in almost all real life applications concerning 3D data. While existing works strive to find a single solution to the pose estimation problem, we make peace with the ambiguities causing high uncertainty around which solutions to identify as the best. Instead, we report a family of poses which capture the nature of the solution space. DBN extends the state of the art direct pose regression networks by (i) a multi-hypotheses prediction head which can yield different distribution modes; and (ii) novel loss functions that benefit from Bingham distributions on rotations. This way, DBN can work both in unambiguous cases providing uncertainty information, and in ambiguous scenes where an uncertainty per mode is desired. On a technical front, our network regresses continuous Bingham mixture models and is applicable to both 2D data such as images and to 3D data such as point clouds. We proposed new training strategies so as to avoid mode or posterior collapse during training and to improve numerical stability. Our methods are thoroughly tested on two different applications exploiting two different modalities: (i) 6D camera relocalization from images; and (ii) object pose estimation from 3D point clouds, demonstrating decent advantages over the state of the art. For the former we contributed our own dataset composed of five indoor scenes where it is unavoidable to capture images corresponding to views that are hard to uniquely identify. For the latter we achieve the top results especially for symmetric objects of ModelNet dataset (Wu et al., in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1912–1920, 2015). The code and dataset accompanying this paper is provided under https://multimodal3dvision.github.io.
Involved external institutions
Technische Universität München (TUM)
Germany (DE)
Stanford University
United States (USA) (US)
Siemens AG, Sector Corporate Technology
Germany (DE)
Germany (DE)
Stanford University
United States (USA) (US)
Siemens AG, Sector Corporate Technology
Germany (DE)
How to cite
APA:
Deng, H., Bui, M., Navab, N., Guibas, L., Ilic, S., & Birdal, T. (2022). Deep Bingham Networks: Dealing with Uncertainty and Ambiguity in Pose Estimation. International Journal of Computer Vision, 130(7), 1627-1654. https://doi.org/10.1007/s11263-022-01612-w
MLA:
Deng, Haowen, et al. "Deep Bingham Networks: Dealing with Uncertainty and Ambiguity in Pose Estimation." International Journal of Computer Vision 130.7 (2022): 1627-1654.
BibTeX: Download