This paper proposes a LIDAR-based pedestrian detection method using 3DCNN. The proposed method converts a sparse point-cloud obtained by a low-resolution LIDAR to two-channels voxel representation that consists of the 3D object probability channel and the reflection intensity channel. To evaluate the performance of the proposed method, an experiment using real-world LIDAR data was conducted. The results show that the proposed method is able to detect pedestrians more accurately than detectors trained by other conventional features.
Description
Pedestrian detection from sparse point-cloud using 3DCNN - IEEE Conference Publication
%0 Conference Paper
%1 8369680
%A Tatebe, Y.
%A Deguchi, D.
%A Kawanishi, Y.
%A Ide, I.
%A Murase, H.
%A Sakai, U.
%B 2018 International Workshop on Advanced Image Technology (IWAIT)
%D 2018
%K dnn lidar order1 pedestrian
%P 1-4
%R 10.1109/IWAIT.2018.8369680
%T Pedestrian detection from sparse point-cloud using 3DCNN
%U https://ieeexplore.ieee.org/abstract/document/8369680
%X This paper proposes a LIDAR-based pedestrian detection method using 3DCNN. The proposed method converts a sparse point-cloud obtained by a low-resolution LIDAR to two-channels voxel representation that consists of the 3D object probability channel and the reflection intensity channel. To evaluate the performance of the proposed method, an experiment using real-world LIDAR data was conducted. The results show that the proposed method is able to detect pedestrians more accurately than detectors trained by other conventional features.
@inproceedings{8369680,
abstract = {This paper proposes a LIDAR-based pedestrian detection method using 3DCNN. The proposed method converts a sparse point-cloud obtained by a low-resolution LIDAR to two-channels voxel representation that consists of the 3D object probability channel and the reflection intensity channel. To evaluate the performance of the proposed method, an experiment using real-world LIDAR data was conducted. The results show that the proposed method is able to detect pedestrians more accurately than detectors trained by other conventional features.},
added-at = {2020-05-27T23:38:46.000+0200},
author = {{Tatebe}, Y. and {Deguchi}, D. and {Kawanishi}, Y. and {Ide}, I. and {Murase}, H. and {Sakai}, U.},
biburl = {https://www.bibsonomy.org/bibtex/20205912371af68db19562a13839f7c54/sohnki},
booktitle = {2018 International Workshop on Advanced Image Technology (IWAIT)},
description = {Pedestrian detection from sparse point-cloud using 3DCNN - IEEE Conference Publication},
doi = {10.1109/IWAIT.2018.8369680},
interhash = {165bbb4da569c2a8ee0d6be1f8062c65},
intrahash = {0205912371af68db19562a13839f7c54},
keywords = {dnn lidar order1 pedestrian},
month = jan,
pages = {1-4},
timestamp = {2020-06-02T20:02:09.000+0200},
title = {Pedestrian detection from sparse point-cloud using 3DCNN},
url = {https://ieeexplore.ieee.org/abstract/document/8369680},
year = 2018
}