Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans

Kazuto Nakashima; Seungwoo Nham; Hojung Jung; Yumi Iwashita; Ryo Kurazume; Oscar M. Mozos

doi:10.1109/SII.2017.8279243

Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans

Kazuto Nakashima, Seungwoo Nham, Hojung Jung, Yumi Iwashita, Ryo Kurazume, Oscar M. Mozos

Real world robotics

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

Abstract

We present a novel method for the outdoor scene categorization using 2D convolutional neural networks (CNNs) which take panoramic depth images obtained by a 3D laser scanner as input. We evaluate our approach in two outdoor scene datasets including six categories: coast, forest, indoor parking, outdoor parking, residential area, and urban area. Our results on both datasets (over 94%) outperform previous approaches and show the effectiveness of this approach for outdoor scene categorization using depth images. To analyze our trained networks we visualize the learned features by using two visualization methods.

Original language	English
Title of host publication	SII 2017 - 2017 IEEE/SICE International Symposium on System Integration
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	387-392
Number of pages	6
Volume	2018-January
ISBN (Electronic)	9781538622636
DOIs	https://doi.org/10.1109/SII.2017.8279243
Publication status	Published - Feb 1 2018
Event	2017 IEEE/SICE International Symposium on System Integration, SII 2017 - Taipei, Taiwan, Province of China Duration: Dec 11 2017 → Dec 14 2017

Publication series

Name	SII 2017 - 2017 IEEE/SICE International Symposium on System Integration
Volume	2018-January

Conference

Conference	2017 IEEE/SICE International Symposium on System Integration, SII 2017
Country/Territory	Taiwan, Province of China
City	Taipei
Period	12/11/17 → 12/14/17

All Science Journal Classification (ASJC) codes

Modelling and Simulation
Instrumentation
Artificial Intelligence
Computer Science Applications
Engineering (miscellaneous)
Materials Science (miscellaneous)
Control and Optimization

Access to Document

10.1109/SII.2017.8279243

Cite this

Nakashima, K., Nham, S., Jung, H., Iwashita, Y., Kurazume, R., & Mozos, O. M. (2018). Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans. In SII 2017 - 2017 IEEE/SICE International Symposium on System Integration (Vol. 2018-January, pp. 387-392). (SII 2017 - 2017 IEEE/SICE International Symposium on System Integration; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SII.2017.8279243

Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans. / Nakashima, Kazuto; Nham, Seungwoo; Jung, Hojung et al.
SII 2017 - 2017 IEEE/SICE International Symposium on System Integration. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. p. 387-392 (SII 2017 - 2017 IEEE/SICE International Symposium on System Integration; Vol. 2018-January).

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

Nakashima, K, Nham, S, Jung, H, Iwashita, Y, Kurazume, R & Mozos, OM 2018, Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans. in SII 2017 - 2017 IEEE/SICE International Symposium on System Integration. vol. 2018-January, SII 2017 - 2017 IEEE/SICE International Symposium on System Integration, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 387-392, 2017 IEEE/SICE International Symposium on System Integration, SII 2017, Taipei, Taiwan, Province of China, 12/11/17. https://doi.org/10.1109/SII.2017.8279243

Nakashima K, Nham S, Jung H, Iwashita Y, Kurazume R, Mozos OM. Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans. In SII 2017 - 2017 IEEE/SICE International Symposium on System Integration. Vol. 2018-January. Institute of Electrical and Electronics Engineers Inc. 2018. p. 387-392. (SII 2017 - 2017 IEEE/SICE International Symposium on System Integration). doi: 10.1109/SII.2017.8279243

Nakashima, Kazuto ; Nham, Seungwoo ; Jung, Hojung et al. / Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans. SII 2017 - 2017 IEEE/SICE International Symposium on System Integration. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. pp. 387-392 (SII 2017 - 2017 IEEE/SICE International Symposium on System Integration).

@inproceedings{206248efbd694855a21d936937d426ad,

title = "Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans",

abstract = "We present a novel method for the outdoor scene categorization using 2D convolutional neural networks (CNNs) which take panoramic depth images obtained by a 3D laser scanner as input. We evaluate our approach in two outdoor scene datasets including six categories: coast, forest, indoor parking, outdoor parking, residential area, and urban area. Our results on both datasets (over 94%) outperform previous approaches and show the effectiveness of this approach for outdoor scene categorization using depth images. To analyze our trained networks we visualize the learned features by using two visualization methods.",

author = "Kazuto Nakashima and Seungwoo Nham and Hojung Jung and Yumi Iwashita and Ryo Kurazume and Mozos, {Oscar M.}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 IEEE/SICE International Symposium on System Integration, SII 2017 ; Conference date: 11-12-2017 Through 14-12-2017",

year = "2018",

month = feb,

day = "1",

doi = "10.1109/SII.2017.8279243",

language = "English",

volume = "2018-January",

series = "SII 2017 - 2017 IEEE/SICE International Symposium on System Integration",

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

pages = "387--392",

booktitle = "SII 2017 - 2017 IEEE/SICE International Symposium on System Integration",

address = "United States",

}

TY - GEN

T1 - Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans

AU - Nakashima, Kazuto

AU - Nham, Seungwoo

AU - Jung, Hojung

AU - Iwashita, Yumi

AU - Kurazume, Ryo

AU - Mozos, Oscar M.

PY - 2018/2/1

Y1 - 2018/2/1

N2 - We present a novel method for the outdoor scene categorization using 2D convolutional neural networks (CNNs) which take panoramic depth images obtained by a 3D laser scanner as input. We evaluate our approach in two outdoor scene datasets including six categories: coast, forest, indoor parking, outdoor parking, residential area, and urban area. Our results on both datasets (over 94%) outperform previous approaches and show the effectiveness of this approach for outdoor scene categorization using depth images. To analyze our trained networks we visualize the learned features by using two visualization methods.

AB - We present a novel method for the outdoor scene categorization using 2D convolutional neural networks (CNNs) which take panoramic depth images obtained by a 3D laser scanner as input. We evaluate our approach in two outdoor scene datasets including six categories: coast, forest, indoor parking, outdoor parking, residential area, and urban area. Our results on both datasets (over 94%) outperform previous approaches and show the effectiveness of this approach for outdoor scene categorization using depth images. To analyze our trained networks we visualize the learned features by using two visualization methods.

UR - http://www.scopus.com/inward/record.url?scp=85050869408&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050869408&partnerID=8YFLogxK

U2 - 10.1109/SII.2017.8279243

DO - 10.1109/SII.2017.8279243

M3 - Conference contribution

AN - SCOPUS:85050869408

VL - 2018-January

T3 - SII 2017 - 2017 IEEE/SICE International Symposium on System Integration

SP - 387

EP - 392

BT - SII 2017 - 2017 IEEE/SICE International Symposium on System Integration

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE/SICE International Symposium on System Integration, SII 2017

Y2 - 11 December 2017 through 14 December 2017

ER -

Recognizing outdoor scenes by convolutional features of omni-directional LiDAR scans

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this