Road Lane Semantic Segmentation for High Definition Map

Wonje Jang; Jhonghyun An; Sangyun Lee; Minho Cho; Myungki Sun; Euntai Kim

doi:10.1109/IVS.2018.8500661

Road Lane Semantic Segmentation for High Definition Map

Wonje Jang, Jhonghyun An, Sangyun Lee, Minho Cho, Myungki Sun, Euntai Kim

Department of Electrical and Electronic Engineering

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

12 Citations (Scopus)

Abstract

High Definition map (HD Map) is an important part of autonomous driving vehicle. Most conventional method to generate HD map requires expensive system and postprocessing of observed data. In this paper, we propose automatic HD map generating algorithm using just monocular camera without further human labors. The proposed algorithm detects road lane from image and classifies the type of road lane at pixel-level with Fully Convolutional Network (FCN) which outperforms the other semantic segmentation methods. The segmentation results are used to extract lane features, and the features are used for loop-closure detection. Final map is generated with graph-based Simultaneous Localization and Mapping (SLAM) algorithm. The experiment is done with monocular camera mounted on mobile vehicle. In this paper, final map generated by proposed method is compared with aerial view data. The results show that the proposed method can generate reliable map that is comparable to real roads even only the low-cost sensor is used.

Original language	English
Title of host publication	2018 IEEE Intelligent Vehicles Symposium, IV 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1001-1006
Number of pages	6
ISBN (Electronic)	9781538644522
DOIs	https://doi.org/10.1109/IVS.2018.8500661
Publication status	Published - 2018 Oct 18
Event	2018 IEEE Intelligent Vehicles Symposium, IV 2018 - Changshu, Suzhou, China Duration: 2018 Sept 26 → 2018 Sept 30

Publication series

Name	IEEE Intelligent Vehicles Symposium, Proceedings
Volume	2018-June

Other

Other	2018 IEEE Intelligent Vehicles Symposium, IV 2018
Country/Territory	China
City	Changshu, Suzhou
Period	18/9/26 → 18/9/30

Bibliographical note

Funding Information:
W. Jang, J. An, S. Lee, M. Cho and E. Kim* is in the School of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Korea (email: [jangwj1256; jhonghyen; hello072; minho8849; etkim] @yonsei.ac.kr) M. Sun is in Hyundai Mnsoft, Hyundai Motor bldg., 74, Wonhyo-ro, Yongsan-gu, Seoul 04365, Korea (email: mk77@hyundai-mnsoft.com) This research was supported by ‘Machine Learning-based Development of Lane Extraction Algorithm from Mobile Mapping System(MMS)' project funded by the HYUNDAI-MnSOFT Corporation. *Corresponding author

Publisher Copyright:
© 2018 IEEE.

All Science Journal Classification (ASJC) codes

Computer Science Applications
Automotive Engineering
Modelling and Simulation

Access to Document

10.1109/IVS.2018.8500661

Cite this

@inproceedings{353f2fd19c074ddabb7fac5fd3a22b4d,

title = "Road Lane Semantic Segmentation for High Definition Map",

abstract = "High Definition map (HD Map) is an important part of autonomous driving vehicle. Most conventional method to generate HD map requires expensive system and postprocessing of observed data. In this paper, we propose automatic HD map generating algorithm using just monocular camera without further human labors. The proposed algorithm detects road lane from image and classifies the type of road lane at pixel-level with Fully Convolutional Network (FCN) which outperforms the other semantic segmentation methods. The segmentation results are used to extract lane features, and the features are used for loop-closure detection. Final map is generated with graph-based Simultaneous Localization and Mapping (SLAM) algorithm. The experiment is done with monocular camera mounted on mobile vehicle. In this paper, final map generated by proposed method is compared with aerial view data. The results show that the proposed method can generate reliable map that is comparable to real roads even only the low-cost sensor is used.",

author = "Wonje Jang and Jhonghyun An and Sangyun Lee and Minho Cho and Myungki Sun and Euntai Kim",

note = "Funding Information: W. Jang, J. An, S. Lee, M. Cho and E. Kim* is in the School of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Korea (email: [jangwj1256; jhonghyen; hello072; minho8849; etkim] @yonsei.ac.kr) M. Sun is in Hyundai Mnsoft, Hyundai Motor bldg., 74, Wonhyo-ro, Yongsan-gu, Seoul 04365, Korea (email: mk77@hyundai-mnsoft.com) This research was supported by {\textquoteleft}Machine Learning-based Development of Lane Extraction Algorithm from Mobile Mapping System(MMS)' project funded by the HYUNDAI-MnSOFT Corporation. *Corresponding author Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Intelligent Vehicles Symposium, IV 2018 ; Conference date: 26-09-2018 Through 30-09-2018",

year = "2018",

month = oct,

day = "18",

doi = "10.1109/IVS.2018.8500661",

language = "English",

series = "IEEE Intelligent Vehicles Symposium, Proceedings",

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

pages = "1001--1006",

booktitle = "2018 IEEE Intelligent Vehicles Symposium, IV 2018",

address = "United States",

}

Jang, W, An, J, Lee, S, Cho, M, Sun, M & Kim, E 2018, Road Lane Semantic Segmentation for High Definition Map. in 2018 IEEE Intelligent Vehicles Symposium, IV 2018., 8500661, IEEE Intelligent Vehicles Symposium, Proceedings, vol. 2018-June, Institute of Electrical and Electronics Engineers Inc., pp. 1001-1006, 2018 IEEE Intelligent Vehicles Symposium, IV 2018, Changshu, Suzhou, China, 18/9/26. https://doi.org/10.1109/IVS.2018.8500661

Road Lane Semantic Segmentation for High Definition Map. / Jang, Wonje; An, Jhonghyun; Lee, Sangyun et al.

2018 IEEE Intelligent Vehicles Symposium, IV 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1001-1006 8500661 (IEEE Intelligent Vehicles Symposium, Proceedings; Vol. 2018-June).

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

TY - GEN

T1 - Road Lane Semantic Segmentation for High Definition Map

AU - Jang, Wonje

AU - An, Jhonghyun

AU - Lee, Sangyun

AU - Cho, Minho

AU - Sun, Myungki

AU - Kim, Euntai

N1 - Funding Information: W. Jang, J. An, S. Lee, M. Cho and E. Kim* is in the School of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Korea (email: [jangwj1256; jhonghyen; hello072; minho8849; etkim] @yonsei.ac.kr) M. Sun is in Hyundai Mnsoft, Hyundai Motor bldg., 74, Wonhyo-ro, Yongsan-gu, Seoul 04365, Korea (email: mk77@hyundai-mnsoft.com) This research was supported by ‘Machine Learning-based Development of Lane Extraction Algorithm from Mobile Mapping System(MMS)' project funded by the HYUNDAI-MnSOFT Corporation. *Corresponding author Publisher Copyright: © 2018 IEEE.

PY - 2018/10/18

Y1 - 2018/10/18

N2 - High Definition map (HD Map) is an important part of autonomous driving vehicle. Most conventional method to generate HD map requires expensive system and postprocessing of observed data. In this paper, we propose automatic HD map generating algorithm using just monocular camera without further human labors. The proposed algorithm detects road lane from image and classifies the type of road lane at pixel-level with Fully Convolutional Network (FCN) which outperforms the other semantic segmentation methods. The segmentation results are used to extract lane features, and the features are used for loop-closure detection. Final map is generated with graph-based Simultaneous Localization and Mapping (SLAM) algorithm. The experiment is done with monocular camera mounted on mobile vehicle. In this paper, final map generated by proposed method is compared with aerial view data. The results show that the proposed method can generate reliable map that is comparable to real roads even only the low-cost sensor is used.

AB - High Definition map (HD Map) is an important part of autonomous driving vehicle. Most conventional method to generate HD map requires expensive system and postprocessing of observed data. In this paper, we propose automatic HD map generating algorithm using just monocular camera without further human labors. The proposed algorithm detects road lane from image and classifies the type of road lane at pixel-level with Fully Convolutional Network (FCN) which outperforms the other semantic segmentation methods. The segmentation results are used to extract lane features, and the features are used for loop-closure detection. Final map is generated with graph-based Simultaneous Localization and Mapping (SLAM) algorithm. The experiment is done with monocular camera mounted on mobile vehicle. In this paper, final map generated by proposed method is compared with aerial view data. The results show that the proposed method can generate reliable map that is comparable to real roads even only the low-cost sensor is used.

UR - http://www.scopus.com/inward/record.url?scp=85056749916&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056749916&partnerID=8YFLogxK

U2 - 10.1109/IVS.2018.8500661

DO - 10.1109/IVS.2018.8500661

M3 - Conference contribution

AN - SCOPUS:85056749916

T3 - IEEE Intelligent Vehicles Symposium, Proceedings

SP - 1001

EP - 1006

BT - 2018 IEEE Intelligent Vehicles Symposium, IV 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE Intelligent Vehicles Symposium, IV 2018

Y2 - 26 September 2018 through 30 September 2018

ER -

Road Lane Semantic Segmentation for High Definition Map

Abstract

Publication series

Other

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this