Loop-Closure Detection Using Local Relative Orientation Matching

Loop-closure detection (LCD), which aims to recognize a previously visited location, is a crucial component of the simultaneous localization and mapping system. In this paper, a novel appearance-based LCD method is presented. In particular, we propose a simple yet surprisingly useful feature matchin...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on intelligent transportation systems Ročník 23; číslo 7; s. 7896 - 7909
Hlavní autoři: Ma, Jiayi, Ye, Xinyu, Zhou, Huabing, Mei, Xiaoguang, Fan, Fan
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.07.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algorithms
ASMK
Feature extraction
feature matching
Liquid crystal displays
Loop-closure detection
Matching
Neural networks
place recognition
Real-time systems
Simultaneous localization and mapping
SLAM
Task analysis
Visualization
ISSN:1524-9050, 1558-0016
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Loop-closure detection (LCD), which aims to recognize a previously visited location, is a crucial component of the simultaneous localization and mapping system. In this paper, a novel appearance-based LCD method is presented. In particular, we propose a simple yet surprisingly useful feature matching algorithm for real-time geometrical verification of candidate loop-closures, termed as local relative orientation matching (LRO). It aims to efficiently establish reliable feature correspondences based on preserving local topological structures between the query image and candidate frame. To effectively retrieve candidate loop closures, we introduce the aggregated selective match kernel framework into the LCD task, which can effectively represent images and reduce the quantization noise of the traditional bag-of-words framework. In addition, the SuperPoint neural network is employed to extract reliable interest points and feature descriptors. Extensive experimental results demonstrate that our LRO can significantly improve the LCD performance, and the proposed overall LCD method can achieve much better performance over the current state-of-the-art on six publicly available datasets.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1524-9050
1558-0016
DOI:10.1109/TITS.2021.3074520