Skip to main navigation Skip to search Skip to main content

Point Cloud Registration Using Evolutionary Algorithm

  • Xi'an Jiaotong University

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

Abstract

Point cloud registration is a hot research topic in computer vision and related areas. In this paper, inspired by Iterative Closest Point (ICP), two evolutionary algorithms, that is, Genetic Algorithm (GA) and Estimation of Distribution Algorithm (EDA), are designed to deal with the problem of registration. Our study shows that evolutionary algorithm is potential to be applied to this problem, it can obtain better results than ICP on the tested real-world datum. Moreover, EDA can provide better results than GA.

Original languageEnglish
Title of host publicationBio-Inspired Computing
Subtitle of host publicationTheories and Applications - 15th International Conference, BIC-TA 2020, Revised Selected Papers
EditorsLinqiang Pan, Shangchen Pang, Tao Song, Faming Gong
PublisherSpringer Science and Business Media Deutschland GmbH
Pages69-77
Number of pages9
ISBN (Print)9789811613531
DOIs
StatePublished - 2021
Event15th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2020 - Qingdao, China
Duration: 23 Oct 202025 Oct 2020

Publication series

NameCommunications in Computer and Information Science
Volume1363 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference15th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2020
Country/TerritoryChina
CityQingdao
Period23/10/2025/10/20

Keywords

  • Estimation of distribution algorithm
  • Evolutionary algorithm
  • Genetic algorithm
  • Point cloud registration

Fingerprint

Dive into the research topics of 'Point Cloud Registration Using Evolutionary Algorithm'. Together they form a unique fingerprint.

Cite this