LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis

Zhe Liu; Shunbo Zhou; Chuanzhe Suo; Peng Yin; Wen Chen; Hesheng Wang; Haoang Li; Yun-Hui Liu

doi:10.1109/ICCV.2019.00292

LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis

Zhe Liu
, Shunbo Zhou
, Chuanzhe Suo
, Peng Yin
, Wen Chen
, Hesheng Wang^*
, Haoang Li
, Yun-Hui Liu

^*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

337 Citations (Scopus)

Abstract

Point cloud based place recognition is still an open issue due to the difficulty in extracting local features from the raw 3D point cloud and generating the global descriptor, and it's even harder in the large-scale dynamic environments. In this paper, we develop a novel deep neural network, named LPD-Net (Large-scale Place Description Network), which can extract discriminative and generalizable global descriptors from the raw 3D point cloud. Two modules, the adaptive local feature extraction module and the graph-based neighborhood aggregation module, are proposed, which contribute to extract the local structures and reveal the spatial distribution of local features in the large-scale point cloud, with an end-to-end manner. We implement the proposed global descriptor in solving point cloud based retrieval tasks to achieve the large-scale place recognition. Comparison results show that our LPD-Net is much better than PointNetVLAD and reaches the state-of-the-art. We also compare our LPD-Net with the vision-based solutions to show the robustness of our approach to different weather and light conditions. © 2019 IEEE.

Original language	English
Title of host publication	Proceedings - 2019 International Conference on Computer Vision, ICCV 2019
Publisher	IEEE
Pages	2831-2840
ISBN (Electronic)	978-1-7281-4804-5
ISBN (Print)	978-1-7281-4803-8
DOIs	https://doi.org/10.1109/ICCV.2019.00292
Publication status	Published - Oct 2019
Externally published	Yes
Event	17th IEEE/CVF International Conference on Computer Vision (ICCV 2019) - COEX Convention Center, Seoul, Korea, Republic of Duration: 27 Oct 2019 → 2 Nov 2019 http://iccv2019.thecvf.com/

Publication series

Name	Proceedings of the IEEE International Conference on Computer Vision
ISSN (Print)	1550-5499
ISSN (Electronic)	2380-7504

Conference

Conference	17th IEEE/CVF International Conference on Computer Vision (ICCV 2019)
Abbreviated title	ICCV19
Place	Korea, Republic of
City	Seoul
Period	27/10/19 → 2/11/19
Internet address	http://iccv2019.thecvf.com/

Funding

This work is supported in part by the Natural Science Foundation of China under Grant U1613218, in part by the Hong Kong ITC under Grant ITS/448/16FP, in part by the National Key Research and Development Program of China under Grant 2018YF- B1309300, and in part by the VC Fund 4930745 of the CUHK T Stone Robotics Institute.

Access Output

10.1109/ICCV.2019.00292

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Liu, Z., Zhou, S., Suo, C., Yin, P., Chen, W., Wang, H., Li, H., & Liu, Y.-H. (2019). LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis. In Proceedings - 2019 International Conference on Computer Vision, ICCV 2019 (pp. 2831-2840). (Proceedings of the IEEE International Conference on Computer Vision). IEEE. https://doi.org/10.1109/ICCV.2019.00292

@inproceedings{728802cf676d416098cc2bf9c20ebd76,

title = "LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis",

abstract = "Point cloud based place recognition is still an open issue due to the difficulty in extracting local features from the raw 3D point cloud and generating the global descriptor, and it's even harder in the large-scale dynamic environments. In this paper, we develop a novel deep neural network, named LPD-Net (Large-scale Place Description Network), which can extract discriminative and generalizable global descriptors from the raw 3D point cloud. Two modules, the adaptive local feature extraction module and the graph-based neighborhood aggregation module, are proposed, which contribute to extract the local structures and reveal the spatial distribution of local features in the large-scale point cloud, with an end-to-end manner. We implement the proposed global descriptor in solving point cloud based retrieval tasks to achieve the large-scale place recognition. Comparison results show that our LPD-Net is much better than PointNetVLAD and reaches the state-of-the-art. We also compare our LPD-Net with the vision-based solutions to show the robustness of our approach to different weather and light conditions. {\textcopyright} 2019 IEEE.",

author = "Zhe Liu and Shunbo Zhou and Chuanzhe Suo and Peng Yin and Wen Chen and Hesheng Wang and Haoang Li and Yun-Hui Liu",

year = "2019",

month = oct,

doi = "10.1109/ICCV.2019.00292",

language = "English",

isbn = "978-1-7281-4803-8",

series = "Proceedings of the IEEE International Conference on Computer Vision",

publisher = "IEEE",

pages = "2831--2840",

booktitle = "Proceedings - 2019 International Conference on Computer Vision, ICCV 2019",

address = "United States",

note = "17th IEEE/CVF International Conference on Computer Vision (ICCV 2019), ICCV19 ; Conference date: 27-10-2019 Through 02-11-2019",

url = "http://iccv2019.thecvf.com/",

}

Liu, Z, Zhou, S, Suo, C, Yin, P, Chen, W, Wang, H, Li, H & Liu, Y-H 2019, LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis. in Proceedings - 2019 International Conference on Computer Vision, ICCV 2019. Proceedings of the IEEE International Conference on Computer Vision, IEEE, pp. 2831-2840, 17th IEEE/CVF International Conference on Computer Vision (ICCV 2019), Seoul, Korea, Republic of, 27/10/19. https://doi.org/10.1109/ICCV.2019.00292

LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis. / Liu, Zhe; Zhou, Shunbo; Suo, Chuanzhe et al.
Proceedings - 2019 International Conference on Computer Vision, ICCV 2019. IEEE, 2019. p. 2831-2840 (Proceedings of the IEEE International Conference on Computer Vision).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - LPD-Net

T2 - 17th IEEE/CVF International Conference on Computer Vision (ICCV 2019)

AU - Liu, Zhe

AU - Zhou, Shunbo

AU - Suo, Chuanzhe

AU - Yin, Peng

AU - Chen, Wen

AU - Wang, Hesheng

AU - Li, Haoang

AU - Liu, Yun-Hui

PY - 2019/10

Y1 - 2019/10

N2 - Point cloud based place recognition is still an open issue due to the difficulty in extracting local features from the raw 3D point cloud and generating the global descriptor, and it's even harder in the large-scale dynamic environments. In this paper, we develop a novel deep neural network, named LPD-Net (Large-scale Place Description Network), which can extract discriminative and generalizable global descriptors from the raw 3D point cloud. Two modules, the adaptive local feature extraction module and the graph-based neighborhood aggregation module, are proposed, which contribute to extract the local structures and reveal the spatial distribution of local features in the large-scale point cloud, with an end-to-end manner. We implement the proposed global descriptor in solving point cloud based retrieval tasks to achieve the large-scale place recognition. Comparison results show that our LPD-Net is much better than PointNetVLAD and reaches the state-of-the-art. We also compare our LPD-Net with the vision-based solutions to show the robustness of our approach to different weather and light conditions. © 2019 IEEE.

AB - Point cloud based place recognition is still an open issue due to the difficulty in extracting local features from the raw 3D point cloud and generating the global descriptor, and it's even harder in the large-scale dynamic environments. In this paper, we develop a novel deep neural network, named LPD-Net (Large-scale Place Description Network), which can extract discriminative and generalizable global descriptors from the raw 3D point cloud. Two modules, the adaptive local feature extraction module and the graph-based neighborhood aggregation module, are proposed, which contribute to extract the local structures and reveal the spatial distribution of local features in the large-scale point cloud, with an end-to-end manner. We implement the proposed global descriptor in solving point cloud based retrieval tasks to achieve the large-scale place recognition. Comparison results show that our LPD-Net is much better than PointNetVLAD and reaches the state-of-the-art. We also compare our LPD-Net with the vision-based solutions to show the robustness of our approach to different weather and light conditions. © 2019 IEEE.

UR - https://www.scopus.com/pages/publications/85081933469

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85081933469&origin=recordpage

U2 - 10.1109/ICCV.2019.00292

DO - 10.1109/ICCV.2019.00292

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 978-1-7281-4803-8

T3 - Proceedings of the IEEE International Conference on Computer Vision

SP - 2831

EP - 2840

BT - Proceedings - 2019 International Conference on Computer Vision, ICCV 2019

PB - IEEE

Y2 - 27 October 2019 through 2 November 2019

ER -

LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis

Abstract

Publication series

Conference

Funding

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this