arrow
返回

PSE-Match: A Viewpoint-Free Place Recognition Method With Parallel Semantic Embedding

delete2022-08-01
delete14
delete
OA
AI
尹鹏 封面图
尹鹏 (Peng Yin)
L
Lingyun Xu *
Z
Ziyue Feng
A
Anton Egorov
B
Bing Li
DOI:10.1109/TITS.2021.3102429delete
delete原文链接
delete分享
delete收藏
摘要

摘要

En 中文
Accurate localization on the autonomous driving cars is essential for autonomy and driving safety, especially for complex urban streets and search-and-rescue subterranean environments where high-accurate GPS is not available. However current odometry estimation may introduce the drifting problems in long-term navigation without robust global localization. The main challenges involve scene divergence under the interference of dynamic environments and effective perception of observation and object layout variance from different viewpoints. To tackle these challenges, we present PSE-Match, a viewpoint-free place recognition method based on parallel semantic analysis of isolated semantic attributes from 3D point-cloud models. Compared with the original point cloud, the observed variance of semantic attributes is smaller. PSE-Match incorporates a divergence place learning network to capture different semantic attributes parallelly through the spherical harmonics domain. Using both existing benchmark datasets and two in-field collected datasets, our experiments show that the proposed method achieves above 70% average recall with top one retrieval and above 95% average recall with top ten retrieval cases. And PSE-Match has also demonstrated an obvious generalization ability with limited training dataset.
Keyword:
Semantics
Feature extraction
Three-dimensional displays
Octrees
Location awareness
Laser radar
Convolution
3D place recognition
semantic embedding
divergence learning
global localization
place feature learning

期刊

IEEE Transactions on Intelligent Transportation Systems 封面图
IEEE Transactions on Intelligent Transportation Systems
IF:
8.4
论文数:
9.7K
被引数:
6.3W

机构

C
Carnegie Mellon University
学者数:
1.4W
论文数: 1.4W
被引数: 2.7W
S
skolkovo institute of science & technology
学者数:
3.3K
论文数: 2.3K
被引数: 1
C
Clemson University
学者数:
1.3W
论文数: 1.1W
被引数: 1.4W
学者 查看更多机构
引用论文

引用论文

SegMap: Segment-based mapping and localization using data-driven descriptors
err2019-07-10
err132
errOAAI
errDube, Renaud; Cramariuc, Andrei; Dugas, Daniel; Sommer, Hannes; Dymczyk, Marcin; Nieto, Juan; Siegwart, Roland; Cadena, Cesar
err分享
err收藏
Thermischer Abbau von Polysulfonen
err1967-10-01
err0
PREAI
errHorst-Dieter R. Schüddemage; Dieter O. Hummel
err分享
err收藏
Mutations in KaiA, a clock protein, extend the period of circadian rhythm in the cyanobacterium Synechococcus elongatus PCC 7942
err2002-09-01
err0
errOAAI
errHideya Nishimura; Keiko Imai; Hidehiko Saito; Makoto Hirai; Yoichi Nakahira; Hisayo Kondo; Hiroshi Hayashi; Takao Kondo; Akiko Tsuruhara
err分享
err收藏
学者 查看更多内容