KeySLAM: Robust RGB-D Camera Tracking Using Adaptive VO and Optimal Key-Frame Selection

Kyung Min Han; Young J. Kim

doi:10.1109/LRA.2020.3026964

KeySLAM: Robust RGB-D Camera Tracking Using Adaptive VO and Optimal Key-Frame Selection

Kyung Min Han, Young J. Kim

Department of Computer Science & Engineering

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

We propose a novel RGB-D camera tracking system that robustly reconstructs hand-held RGB-D camera sequences. The robustness of our system is achieved by two independent features of our method: adaptive visual odometry (VO) and integer programming-based key-frame selection. Our VO method adaptively interpolates the camera motion results of the direct VO (DVO) and the iterative closed point (ICP) to yield more optimal results than existing methods such as Elastic-Fusion. Moreover, our key-frame selection method locates globally optimum key-frames using a comprehensive objective function in a deterministic manner rather than heuristic or experience-based rules that prior methods mostly rely on. As a result, our method can complete reconstruction even if the camera fails to be tracked due to discontinuous camera motions, such as kidnap events, when conventional systems need to backtrack the scene.

Original language	English
Article number	9206050
Pages (from-to)	6940-6947
Number of pages	8
Journal	IEEE Robotics and Automation Letters
Volume	5
Issue number	4
DOIs	https://doi.org/10.1109/LRA.2020.3026964
State	Published - Oct 2020

Bibliographical note

Funding Information:
Manuscript received March 10, 2020; accepted July 9, 2020. Date of publication September 25, 2020; date of current version October 7, 2020. This letter was recommended for publication by Associate Editor L. Paull and Editor S. Behnke upon evaluation of the Reviewers’ comments. This work was supported in part by the ITRC/IITP Program (IITP-2020-0-01460), and in part by the NRF (2018R1A6A3A11049832 and 2017R1A2B3012701) in South Korea. (Corresponding author: Young J. Kim.) The authors are with the Department of Computer Science and Engineering, Ewha Womans University, Seoul 03760, South Korea (e-mail: [email protected]; [email protected]).

Publisher Copyright:
© 2016 IEEE.

Keywords

Mapping
RGB-D Perception
and SLAM

Access to Document

10.1109/LRA.2020.3026964

Cite this

@article{84c4725a38e5464a82b63f7f5f4f17b5,

title = "KeySLAM: Robust RGB-D Camera Tracking Using Adaptive VO and Optimal Key-Frame Selection",

abstract = "We propose a novel RGB-D camera tracking system that robustly reconstructs hand-held RGB-D camera sequences. The robustness of our system is achieved by two independent features of our method: adaptive visual odometry (VO) and integer programming-based key-frame selection. Our VO method adaptively interpolates the camera motion results of the direct VO (DVO) and the iterative closed point (ICP) to yield more optimal results than existing methods such as Elastic-Fusion. Moreover, our key-frame selection method locates globally optimum key-frames using a comprehensive objective function in a deterministic manner rather than heuristic or experience-based rules that prior methods mostly rely on. As a result, our method can complete reconstruction even if the camera fails to be tracked due to discontinuous camera motions, such as kidnap events, when conventional systems need to backtrack the scene. ",

keywords = "Mapping, RGB-D Perception, and SLAM",

author = "Han, {Kyung Min} and Kim, {Young J.}",

note = "Funding Information: Manuscript received March 10, 2020; accepted July 9, 2020. Date of publication September 25, 2020; date of current version October 7, 2020. This letter was recommended for publication by Associate Editor L. Paull and Editor S. Behnke upon evaluation of the Reviewers{\textquoteright} comments. This work was supported in part by the ITRC/IITP Program (IITP-2020-0-01460), and in part by the NRF (2018R1A6A3A11049832 and 2017R1A2B3012701) in South Korea. (Corresponding author: Young J. Kim.) The authors are with the Department of Computer Science and Engineering, Ewha Womans University, Seoul 03760, South Korea (e-mail: [email protected]; [email protected]). Publisher Copyright: {\textcopyright} 2016 IEEE.",

year = "2020",

month = oct,

doi = "10.1109/LRA.2020.3026964",

language = "English",

volume = "5",

pages = "6940--6947",

journal = "IEEE Robotics and Automation Letters",

issn = "2377-3766",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "4",

}

TY - JOUR

T1 - KeySLAM

T2 - Robust RGB-D Camera Tracking Using Adaptive VO and Optimal Key-Frame Selection

AU - Han, Kyung Min

AU - Kim, Young J.

N1 - Funding Information: Manuscript received March 10, 2020; accepted July 9, 2020. Date of publication September 25, 2020; date of current version October 7, 2020. This letter was recommended for publication by Associate Editor L. Paull and Editor S. Behnke upon evaluation of the Reviewers’ comments. This work was supported in part by the ITRC/IITP Program (IITP-2020-0-01460), and in part by the NRF (2018R1A6A3A11049832 and 2017R1A2B3012701) in South Korea. (Corresponding author: Young J. Kim.) The authors are with the Department of Computer Science and Engineering, Ewha Womans University, Seoul 03760, South Korea (e-mail: [email protected]; [email protected]). Publisher Copyright: © 2016 IEEE.

PY - 2020/10

Y1 - 2020/10

N2 - We propose a novel RGB-D camera tracking system that robustly reconstructs hand-held RGB-D camera sequences. The robustness of our system is achieved by two independent features of our method: adaptive visual odometry (VO) and integer programming-based key-frame selection. Our VO method adaptively interpolates the camera motion results of the direct VO (DVO) and the iterative closed point (ICP) to yield more optimal results than existing methods such as Elastic-Fusion. Moreover, our key-frame selection method locates globally optimum key-frames using a comprehensive objective function in a deterministic manner rather than heuristic or experience-based rules that prior methods mostly rely on. As a result, our method can complete reconstruction even if the camera fails to be tracked due to discontinuous camera motions, such as kidnap events, when conventional systems need to backtrack the scene.

AB - We propose a novel RGB-D camera tracking system that robustly reconstructs hand-held RGB-D camera sequences. The robustness of our system is achieved by two independent features of our method: adaptive visual odometry (VO) and integer programming-based key-frame selection. Our VO method adaptively interpolates the camera motion results of the direct VO (DVO) and the iterative closed point (ICP) to yield more optimal results than existing methods such as Elastic-Fusion. Moreover, our key-frame selection method locates globally optimum key-frames using a comprehensive objective function in a deterministic manner rather than heuristic or experience-based rules that prior methods mostly rely on. As a result, our method can complete reconstruction even if the camera fails to be tracked due to discontinuous camera motions, such as kidnap events, when conventional systems need to backtrack the scene.

KW - Mapping

KW - RGB-D Perception

KW - and SLAM

UR - http://www.scopus.com/inward/record.url?scp=85091895260&partnerID=8YFLogxK

U2 - 10.1109/LRA.2020.3026964

DO - 10.1109/LRA.2020.3026964

M3 - Article

AN - SCOPUS:85091895260

SN - 2377-3766

VL - 5

SP - 6940

EP - 6947

JO - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

IS - 4

M1 - 9206050

ER -

KeySLAM: Robust RGB-D Camera Tracking Using Adaptive VO and Optimal Key-Frame Selection

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this