PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation

Arno Solin; Santiago Cortés Reina; Esa Rahtu; Juho Kannala

doi:10.1109/WACV.2018.00073

PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation

Arno Solin^*, Santiago Cortés Reina, Esa Rahtu, Juho Kannala

^*Corresponding author for this work

Tampere University of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

29 Citations (Scopus)

278 Downloads (Pure)

Abstract

This paper presents a novel method for visual-inertial odometry. The method is based on an information fusion framework employing low-cost IMU sensors and the monocular camera in a standard smartphone. We formulate a sequential inference scheme, where the IMU drives the dynamical model and the camera frames are used in coupling trailing sequences of augmented poses. The novelty in the model is in taking into account all the cross-terms in the updates, thus propagating the inter-connected uncertainties throughout the model. Stronger coupling between the inertial and visual data sources leads to robustness against occlusion and feature-poor environments. We demonstrate results on data collected with an iPhone and provide comparisons against the Tango device and using the EuRoC data set.

Original language	English
Title of host publication	Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018
Publisher	IEEE
Pages	616-625
Number of pages	10
ISBN (Electronic)	9781538648865
DOIs	https://doi.org/10.1109/WACV.2018.00073
Publication status	Published - 2018
MoE publication type	A4 Conference publication
Event	IEEE Winter Conference on Applications of Computer Vision - New York, United States Duration: 12 Mar 2018 → 15 Mar 2018 Conference number: 18

Publication series

Name	IEEE Winter Conference on Applications of Computer Vision
Publisher	IEEE
ISSN (Print)	2472-6737

Conference

Conference	IEEE Winter Conference on Applications of Computer Vision
Abbreviated title	WACV
Country/Territory	United States
City	New York
Period	12/03/2018 → 15/03/2018

Keywords

TRACKING
CAMERA

Access to Document

10.1109/WACV.2018.00073

SCI_Solin-et-al-PIVO-Probabilistic_WACVAccepted author manuscript, 3.57 MB

https://arxiv.org/abs/1708.00894

Sequential inference for real-time probabilistic modelling
Solin, A. (Principal investigator)
01/09/2017 → 31/08/2020
Project: Academy of Finland: Other research funding
Crowdsourced mapping of the environment- multimodal real-time SLAM via combined inertial, optical, and magnetic sensoring
Cortés Reina, S. (Project Member), Kannala, J. (Principal investigator), Laskar, Z. (Project Member), Ylioinas, J. (Project Member) & Melekhov, I. (Project Member)
01/01/2016 → 31/12/2017
Project: Academy of Finland: Other research funding

Cite this

@inproceedings{614d5640cb70452bac2d25d02e3d3c92,

title = "PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation",

abstract = "This paper presents a novel method for visual-inertial odometry. The method is based on an information fusion framework employing low-cost IMU sensors and the monocular camera in a standard smartphone. We formulate a sequential inference scheme, where the IMU drives the dynamical model and the camera frames are used in coupling trailing sequences of augmented poses. The novelty in the model is in taking into account all the cross-terms in the updates, thus propagating the inter-connected uncertainties throughout the model. Stronger coupling between the inertial and visual data sources leads to robustness against occlusion and feature-poor environments. We demonstrate results on data collected with an iPhone and provide comparisons against the Tango device and using the EuRoC data set.",

keywords = "TRACKING, CAMERA",

author = "Arno Solin and {Cort{\'e}s Reina}, Santiago and Esa Rahtu and Juho Kannala",

year = "2018",

doi = "10.1109/WACV.2018.00073",

language = "English",

series = "IEEE Winter Conference on Applications of Computer Vision",

publisher = "IEEE",

pages = "616--625",

booktitle = "Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018",

address = "United States",

note = "IEEE Winter Conference on Applications of Computer Vision, WACV ; Conference date: 12-03-2018 Through 15-03-2018",

}

Solin, A, Cortés Reina, S, Rahtu, E & Kannala, J 2018, PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation. in Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018. IEEE Winter Conference on Applications of Computer Vision, IEEE, pp. 616-625, IEEE Winter Conference on Applications of Computer Vision, New York, New York, United States, 12/03/2018. https://doi.org/10.1109/WACV.2018.00073

PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation. / Solin, Arno; Cortés Reina, Santiago; Rahtu, Esa et al.
Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018. IEEE, 2018. p. 616-625 (IEEE Winter Conference on Applications of Computer Vision).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - PIVO

T2 - IEEE Winter Conference on Applications of Computer Vision

AU - Solin, Arno

AU - Cortés Reina, Santiago

AU - Rahtu, Esa

AU - Kannala, Juho

N1 - Conference code: 18

PY - 2018

Y1 - 2018

N2 - This paper presents a novel method for visual-inertial odometry. The method is based on an information fusion framework employing low-cost IMU sensors and the monocular camera in a standard smartphone. We formulate a sequential inference scheme, where the IMU drives the dynamical model and the camera frames are used in coupling trailing sequences of augmented poses. The novelty in the model is in taking into account all the cross-terms in the updates, thus propagating the inter-connected uncertainties throughout the model. Stronger coupling between the inertial and visual data sources leads to robustness against occlusion and feature-poor environments. We demonstrate results on data collected with an iPhone and provide comparisons against the Tango device and using the EuRoC data set.

AB - This paper presents a novel method for visual-inertial odometry. The method is based on an information fusion framework employing low-cost IMU sensors and the monocular camera in a standard smartphone. We formulate a sequential inference scheme, where the IMU drives the dynamical model and the camera frames are used in coupling trailing sequences of augmented poses. The novelty in the model is in taking into account all the cross-terms in the updates, thus propagating the inter-connected uncertainties throughout the model. Stronger coupling between the inertial and visual data sources leads to robustness against occlusion and feature-poor environments. We demonstrate results on data collected with an iPhone and provide comparisons against the Tango device and using the EuRoC data set.

KW - TRACKING

KW - CAMERA

U2 - 10.1109/WACV.2018.00073

DO - 10.1109/WACV.2018.00073

M3 - Conference article in proceedings

T3 - IEEE Winter Conference on Applications of Computer Vision

SP - 616

EP - 625

BT - Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018

PB - IEEE

Y2 - 12 March 2018 through 15 March 2018

ER -

PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

Sequential inference for real-time probabilistic modelling

Crowdsourced mapping of the environment- multimodal real-time SLAM via combined inertial, optical, and magnetic sensoring

Cite this