Omnidirectional DSO: Direct Sparse Odometry with Fisheye Cameras

Matsuki H, Von Stumberg L, Usenko V, Stueckler J, Cremers D (2018)

Publication Type: Journal article

Publication year: 2018

Journal

IEEE Robotics and Automation Letters IEEE

Book Volume: 3

Pages Range: 3693-3700

Article Number: 8410468

Journal Issue: 4

DOI: 10.1109/LRA.2018.2855443

Abstract

We propose a novel real-time direct monocular visual odometry for omnidirectional cameras. Our method extends direct sparse odometry by using the unified omnidirectional model as a projection function, which can be applied to fisheye cameras with a field-of-view (FoV) well above 180$^\circ$. This formulation allows for using the full area of the input image even with strong distortion, while most existing visual odometry methods can only use a rectified and cropped part of it. Model parameters within an active keyframe window are jointly optimized, including the intrinsic/extrinsic camera parameters, three-dimensional position of points, and affine brightness parameters. Thanks to the wide FoV, image overlap between frames becomes bigger and points are more spatially distributed. Our results demonstrate that our method provides increased accuracy and robustness over state-of-the-art visual odometry algorithms.

Involved external institutions

University of Tokyo JP Japan (JP) Technische Universität München (TUM) DE Germany (DE)

How to cite

APA:

Matsuki, H., Von Stumberg, L., Usenko, V., Stueckler, J., & Cremers, D. (2018). Omnidirectional DSO: Direct Sparse Odometry with Fisheye Cameras. IEEE Robotics and Automation Letters, 3(4), 3693-3700. https://doi.org/10.1109/LRA.2018.2855443

MLA:

Matsuki, Hidenobu, et al. "Omnidirectional DSO: Direct Sparse Odometry with Fisheye Cameras." IEEE Robotics and Automation Letters 3.4 (2018): 3693-3700.

BibTeX: Download