FPGA-based ORB feature extraction for real-time visual SLAM

Weikang Fang; Yanjun Zhang; Bo Yu; Shaoshan Liu

doi:10.1109/FPT.2017.8280159

FPGA-based ORB feature extraction for real-time visual SLAM

Weikang Fang
, Yanjun Zhang
, Bo Yu
, Shaoshan Liu

School of Cyberspace Science and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

79 Citations (Scopus)

Abstract

Simultaneous Localization And Mapping (SLAM) is the problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it. How to enable SLAM robustly and durably on mobile, or even IoT grade devices, is the main challenge faced by the industry today. The main problems we need to address are: 1.) how to accelerate the SLAM pipeline to meet real-time requirements; and 2.) how to reduce SLAM energy consumption to extend battery life. After delving into the problem, we found out that feature extraction is indeed the bottleneck of performance and energy consumption. Hence, in this paper, we design, implement, and evaluate a hardware ORB feature extractor and prove that our design is a great balance between performance and energy consumption compared with ARM Krait and Intel Core i5.

Original language	English
Title of host publication	2017 International Conference on Field-Programmable Technology, ICFPT 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	275-278
Number of pages	4
ISBN (Electronic)	9781538626559
DOIs	https://doi.org/10.1109/FPT.2017.8280159
Publication status	Published - 2 Jul 2017
Event	16th IEEE International Conference on Field-Programmable Technology, ICFPT 2017 - Melbourne, Australia Duration: 11 Dec 2017 → 13 Dec 2017

Publication series

Name	2017 International Conference on Field-Programmable Technology, ICFPT 2017
Volume	2018-January

Conference

Conference	16th IEEE International Conference on Field-Programmable Technology, ICFPT 2017
Country/Territory	Australia
City	Melbourne
Period	11/12/17 → 13/12/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

FPGA
Feature extraction
ORB
SLAM

Access to Document

10.1109/FPT.2017.8280159

Cite this

Fang, W., Zhang, Y., Yu, B., & Liu, S. (2017). FPGA-based ORB feature extraction for real-time visual SLAM. In 2017 International Conference on Field-Programmable Technology, ICFPT 2017 (pp. 275-278). (2017 International Conference on Field-Programmable Technology, ICFPT 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/FPT.2017.8280159

@inproceedings{80c00f4258fb4b94bd4eb978b7b0d525,

title = "FPGA-based ORB feature extraction for real-time visual SLAM",

abstract = "Simultaneous Localization And Mapping (SLAM) is the problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it. How to enable SLAM robustly and durably on mobile, or even IoT grade devices, is the main challenge faced by the industry today. The main problems we need to address are: 1.) how to accelerate the SLAM pipeline to meet real-time requirements; and 2.) how to reduce SLAM energy consumption to extend battery life. After delving into the problem, we found out that feature extraction is indeed the bottleneck of performance and energy consumption. Hence, in this paper, we design, implement, and evaluate a hardware ORB feature extractor and prove that our design is a great balance between performance and energy consumption compared with ARM Krait and Intel Core i5.",

keywords = "FPGA, Feature extraction, ORB, SLAM",

author = "Weikang Fang and Yanjun Zhang and Bo Yu and Shaoshan Liu",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 16th IEEE International Conference on Field-Programmable Technology, ICFPT 2017 ; Conference date: 11-12-2017 Through 13-12-2017",

year = "2017",

month = jul,

day = "2",

doi = "10.1109/FPT.2017.8280159",

language = "English",

series = "2017 International Conference on Field-Programmable Technology, ICFPT 2017",

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

pages = "275--278",

booktitle = "2017 International Conference on Field-Programmable Technology, ICFPT 2017",

address = "United States",

}

Fang, W, Zhang, Y, Yu, B & Liu, S 2017, FPGA-based ORB feature extraction for real-time visual SLAM. in 2017 International Conference on Field-Programmable Technology, ICFPT 2017. 2017 International Conference on Field-Programmable Technology, ICFPT 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 275-278, 16th IEEE International Conference on Field-Programmable Technology, ICFPT 2017, Melbourne, Australia, 11/12/17. https://doi.org/10.1109/FPT.2017.8280159

FPGA-based ORB feature extraction for real-time visual SLAM. / Fang, Weikang; Zhang, Yanjun; Yu, Bo et al.
2017 International Conference on Field-Programmable Technology, ICFPT 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 275-278 (2017 International Conference on Field-Programmable Technology, ICFPT 2017; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - FPGA-based ORB feature extraction for real-time visual SLAM

AU - Fang, Weikang

AU - Zhang, Yanjun

AU - Yu, Bo

AU - Liu, Shaoshan

PY - 2017/7/2

Y1 - 2017/7/2

N2 - Simultaneous Localization And Mapping (SLAM) is the problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it. How to enable SLAM robustly and durably on mobile, or even IoT grade devices, is the main challenge faced by the industry today. The main problems we need to address are: 1.) how to accelerate the SLAM pipeline to meet real-time requirements; and 2.) how to reduce SLAM energy consumption to extend battery life. After delving into the problem, we found out that feature extraction is indeed the bottleneck of performance and energy consumption. Hence, in this paper, we design, implement, and evaluate a hardware ORB feature extractor and prove that our design is a great balance between performance and energy consumption compared with ARM Krait and Intel Core i5.

AB - Simultaneous Localization And Mapping (SLAM) is the problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it. How to enable SLAM robustly and durably on mobile, or even IoT grade devices, is the main challenge faced by the industry today. The main problems we need to address are: 1.) how to accelerate the SLAM pipeline to meet real-time requirements; and 2.) how to reduce SLAM energy consumption to extend battery life. After delving into the problem, we found out that feature extraction is indeed the bottleneck of performance and energy consumption. Hence, in this paper, we design, implement, and evaluate a hardware ORB feature extractor and prove that our design is a great balance between performance and energy consumption compared with ARM Krait and Intel Core i5.

KW - FPGA

KW - Feature extraction

KW - ORB

KW - SLAM

UR - https://www.scopus.com/pages/publications/85050863114

U2 - 10.1109/FPT.2017.8280159

DO - 10.1109/FPT.2017.8280159

M3 - Conference contribution

AN - SCOPUS:85050863114

T3 - 2017 International Conference on Field-Programmable Technology, ICFPT 2017

SP - 275

EP - 278

BT - 2017 International Conference on Field-Programmable Technology, ICFPT 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 16th IEEE International Conference on Field-Programmable Technology, ICFPT 2017

Y2 - 11 December 2017 through 13 December 2017

ER -

FPGA-based ORB feature extraction for real-time visual SLAM

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this