CoChat
Sign Up
Cited by 64Advanced Neural Network ApplicationsAutonomous Vehicle Technology and SafetyAdvanced Vision and Imaging

LMNet: Real-time Multiclass Object Detection on CPU Using 3D LiDAR

Open access
Minemura, Kazuki, Liau, Hengfui, Monrroy, Abraham, Kato, Shinpei·2018·2018 3rd Asia-Pacific Conference on Intelligent Robot Systems (ACIRS)
DOI 10.1109/acirs.2018.8467245
Open this paper in CoChat — free accountOpen PDF

Abstract

This paper describes an efficient single-stage deep convolutional neural network to detect objects in urban environments, using nothing more than point cloud data. This feature enables our method to work regardless the time of the day and the lighting conditions. The proposed network structure employs dilated convolutions to gradually increase the perceptive field as depth increases, this helps to reduce the computation time. The network input consists of five perspective representations of the unorganized point cloud data. The network outputs an objectness map and the bounding box offset values for each point. Our experiments showed that using reflection, range, and the position on each of the three axes helped to improve the location and orientation of the output bounding box. We carried out quantitative evaluations with the help of the KITTI dataset evaluation server. It achieved the fastest processing speed among the other contenders, making it suitable for real-time applications. We implemented and tested it on a real vehicle with a Velodyne HDL-64 mounted on top of it. We achieved execution times as fast as 150 FPS using desktop GPUs, and up to 10 FPS on a single Intel Core i5 CPU. The deploy implementation is open-sourced and it can be found as a feature branch inside the autonomous driving framework Autoware. Code is available at: https://github.com/CPFL/Autoware/tree/feature/cnn_lidar_detection.

Integrity signals

Registry

Crossref · 10.1109/acirs.2018.8467245

Cites

21 of 21 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Prior multi-sensor fusion works that introduce LiDAR information only as an additional input modality or cascading revision factors seriously neglect the relation between LiDAR data and Image data at the semantic level.

    “Since both these works only introduce LiDAR information as an additional input modality or cascading revision factors, the relation between LiDAR data and Image data at the semantic level is seriously neglected.”

    From MME-YOLO: Multi-Sensor Multi-Level Enhanced YOLO for Robust Vehicle Detection in Traffic Surveillance (2020) · cites this paper as [14]

  • LMNet designs a network for point clouds that incorporates dilated convolution and 3D object regression loss.

    “Then, LMNet [14] specially designs a network for the point clouds, which incorporates the dilated convolution and 3D object regression loss.”

    From MME-YOLO: Multi-Sensor Multi-Level Enhanced YOLO for Robust Vehicle Detection in Traffic Surveillance (2020) · cites this paper as [14]

  • LMNet achieves real-time inferences of the surrounding cars in three-dimensional bounding boxes.

    “Extended with multi-channel representations from the reflection, the range, the forward distance, the side distance, and the height, LMNet [14] achieves the real-time inferences of the surrounding cars in three-dimensional bounding boxes.”

    From MME-YOLO: Multi-Sensor Multi-Level Enhanced YOLO for Robust Vehicle Detection in Traffic Surveillance (2020) · cites this paper as [14]

Cites

Flagged first, then by citations

6 of the 27 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Minemura, K., Liau, H., Monrroy, A., & Kato, S.
Affiliations
  • Intel (Malaysia) (MY)
  • The University of Tokyo (JP)
ORCID
Venue
2018 3rd Asia-Pacific Conference on Intelligent Robot Systems (ACIRS)
Publisher
IEEE
Published
2018-07
Pages
28-34
DOI
10.1109/acirs.2018.8467245
OpenAlex
W2963299370
Cited by
64
References
27
Citations by year
2026: 2 · 2025: 6 · 2024: 5 · 2023: 8 · 2022: 5 · 2021: 13 · 2020: 13 · 2019: 10 · 2018: 2
FWCI
2.86
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
  • Asia-Pacific Conference on Intelligent Robot Systems (ACIRS) (publishedVersion)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • arXiv (Cornell University) (cc-by)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
Keywords
point clouds, 3D LiDAR, one-stage object detection, dilated convolution, real-time detection, CPU-only inference, autonomous driving, urban environment, KITTI dataset, Autoware, bounding box regression

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review