Skip to main navigation Skip to search Skip to main content

A Line Descriptor for SLAM (LDFS)

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Finding feature correspondence is difficult in low textured scenes, which reduces the performance of simultaneous localization and mapping (SLAM). However, lines that are easy to find can solve this problem well. Generally, features cannot be directly matched, it is necessary to perform some calculations on the features to describe the features and obtain descriptors. In this way, the matching of features can be converted into the matching of descriptors. This article mainly introduces a type of line descriptor that is appropriate for SLAM system. The descriptor fully considers the characteristics of the SLAM system and the geometric constrains of line support regions. The advantage of this line descriptor are as follows: (1) It is more suitable for SLAM system; (2) It is invariant to transformation to some degree; (3) It is robust facing extreme scenes.

Original languageEnglish
Title of host publicationAdvances in Guidance, Navigation and Control - Proceedings of 2020 International Conference on Guidance, Navigation and Control, ICGNC 2020
EditorsLiang Yan, Haibin Duan, Xiang Yu
PublisherSpringer Science and Business Media Deutschland GmbH
Pages4581-4593
Number of pages13
ISBN (Print)9789811581540
DOIs
StatePublished - 2022
EventInternational Conference on Guidance, Navigation and Control, ICGNC 2020 - Tianjin, China
Duration: 23 Oct 202025 Oct 2020

Publication series

NameLecture Notes in Electrical Engineering
Volume644 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Guidance, Navigation and Control, ICGNC 2020
Country/TerritoryChina
CityTianjin
Period23/10/2025/10/20

Keywords

  • Descriptor matching
  • Line feature
  • Robust
  • Suitable for SLAM

Fingerprint

Dive into the research topics of 'A Line Descriptor for SLAM (LDFS)'. Together they form a unique fingerprint.

Cite this