Mixture Density-PoseNet and its Application to Monocular Camera-Based Global Localization

Hyunggi Jo; Woosub Lee; Euntai Kim

doi:10.1109/TII.2020.2986086

Mixture Density-PoseNet and its Application to Monocular Camera-Based Global Localization

Hyunggi Jo, Woosub Lee, Euntai Kim

Department of Electrical and Electronic Engineering

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

3 Citations (Scopus)

Abstract

Global localization using a monocular camera is one of the most challenging problems in computer vision and intelligent robotics. In this article, a new deep neural network named Mixture Density (MD)-PoseNet is proposed to address this problem. Unlike existing learning-based global localization methods that return a single guess for the camera pose, MD-PoseNet returns multiple guesses represented in the form of a Gaussian mixture (GM). The key idea of MD-PoseNet is that the network returns the distribution of all probable camera poses instead of the most probable camera pose, and the distribution represents the multiple guesses for the camera pose. The multiple guesses returned by MD-PoseNet are, consequently, exploited in the probabilistic framework of particle filters. Finally, the proposed method is applied to four different environments, and its validity is demonstrated via experiments.

Original language	English
Article number	9061055
Pages (from-to)	388-397
Number of pages	10
Journal	IEEE Transactions on Industrial Informatics
Volume	17
Issue number	1
DOIs	https://doi.org/10.1109/TII.2020.2986086
Publication status	Published - 2021 Jan

Bibliographical note

Funding Information:
Manuscript received December 7, 2019; revised February 15, 2020; accepted March 18, 2020. Date of publication April 8, 2020; date of current version October 23, 2020. This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210. (Corresponding author: Euntai Kim.) HyungGi Jo and Euntai Kim are with the School of Electrical and Electronic Engineering, Yonsei University, Seoul 03722, South Korea (e-mail: hygijo@yonsei.ac.kr; etkim@yonsei.ac.kr).

Funding Information:
This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210.

Publisher Copyright:
© 2005-2012 IEEE.

All Science Journal Classification (ASJC) codes

Control and Systems Engineering
Information Systems
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/TII.2020.2986086

Cite this

@article{563a5d4a902645b6bf41d882633881af,

title = "Mixture Density-PoseNet and its Application to Monocular Camera-Based Global Localization",

abstract = "Global localization using a monocular camera is one of the most challenging problems in computer vision and intelligent robotics. In this article, a new deep neural network named Mixture Density (MD)-PoseNet is proposed to address this problem. Unlike existing learning-based global localization methods that return a single guess for the camera pose, MD-PoseNet returns multiple guesses represented in the form of a Gaussian mixture (GM). The key idea of MD-PoseNet is that the network returns the distribution of all probable camera poses instead of the most probable camera pose, and the distribution represents the multiple guesses for the camera pose. The multiple guesses returned by MD-PoseNet are, consequently, exploited in the probabilistic framework of particle filters. Finally, the proposed method is applied to four different environments, and its validity is demonstrated via experiments.",

author = "Hyunggi Jo and Woosub Lee and Euntai Kim",

note = "Funding Information: Manuscript received December 7, 2019; revised February 15, 2020; accepted March 18, 2020. Date of publication April 8, 2020; date of current version October 23, 2020. This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210. (Corresponding author: Euntai Kim.) HyungGi Jo and Euntai Kim are with the School of Electrical and Electronic Engineering, Yonsei University, Seoul 03722, South Korea (e-mail: hygijo@yonsei.ac.kr; etkim@yonsei.ac.kr). Funding Information: This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210. Publisher Copyright: {\textcopyright} 2005-2012 IEEE.",

year = "2021",

month = jan,

doi = "10.1109/TII.2020.2986086",

language = "English",

volume = "17",

pages = "388--397",

journal = "IEEE Transactions on Industrial Informatics",

issn = "1551-3203",

publisher = "IEEE Computer Society",

number = "1",

}

TY - JOUR

T1 - Mixture Density-PoseNet and its Application to Monocular Camera-Based Global Localization

AU - Jo, Hyunggi

AU - Lee, Woosub

AU - Kim, Euntai

N1 - Funding Information: Manuscript received December 7, 2019; revised February 15, 2020; accepted March 18, 2020. Date of publication April 8, 2020; date of current version October 23, 2020. This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210. (Corresponding author: Euntai Kim.) HyungGi Jo and Euntai Kim are with the School of Electrical and Electronic Engineering, Yonsei University, Seoul 03722, South Korea (e-mail: hygijo@yonsei.ac.kr; etkim@yonsei.ac.kr). Funding Information: This work was supported by the Industry Core Technology Development Project, 20005062, Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space, funded by the Ministry of Trade, Industry & Energy (MOTIE, South Korea). Paper no. TII-19-5210. Publisher Copyright: © 2005-2012 IEEE.

PY - 2021/1

Y1 - 2021/1

N2 - Global localization using a monocular camera is one of the most challenging problems in computer vision and intelligent robotics. In this article, a new deep neural network named Mixture Density (MD)-PoseNet is proposed to address this problem. Unlike existing learning-based global localization methods that return a single guess for the camera pose, MD-PoseNet returns multiple guesses represented in the form of a Gaussian mixture (GM). The key idea of MD-PoseNet is that the network returns the distribution of all probable camera poses instead of the most probable camera pose, and the distribution represents the multiple guesses for the camera pose. The multiple guesses returned by MD-PoseNet are, consequently, exploited in the probabilistic framework of particle filters. Finally, the proposed method is applied to four different environments, and its validity is demonstrated via experiments.

AB - Global localization using a monocular camera is one of the most challenging problems in computer vision and intelligent robotics. In this article, a new deep neural network named Mixture Density (MD)-PoseNet is proposed to address this problem. Unlike existing learning-based global localization methods that return a single guess for the camera pose, MD-PoseNet returns multiple guesses represented in the form of a Gaussian mixture (GM). The key idea of MD-PoseNet is that the network returns the distribution of all probable camera poses instead of the most probable camera pose, and the distribution represents the multiple guesses for the camera pose. The multiple guesses returned by MD-PoseNet are, consequently, exploited in the probabilistic framework of particle filters. Finally, the proposed method is applied to four different environments, and its validity is demonstrated via experiments.

UR - http://www.scopus.com/inward/record.url?scp=85096035455&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85096035455&partnerID=8YFLogxK

U2 - 10.1109/TII.2020.2986086

DO - 10.1109/TII.2020.2986086

M3 - Article

AN - SCOPUS:85096035455

SN - 1551-3203

VL - 17

SP - 388

EP - 397

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

IS - 1

M1 - 9061055

ER -

Mixture Density-PoseNet and its Application to Monocular Camera-Based Global Localization

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this